EP1238309A1

EP1238309A1 - Ultraviolet and vacuum ultraviolet transparent polymer compositions and their uses

Info

Publication number: EP1238309A1
Application number: EP00979193A
Authority: EP
Inventors: Robert Clayton Wheland; Roger Harquail French; Fredrick Claus Zumsteg, Jr.
Original assignee: EI Du Pont de Nemours and Co
Current assignee: EIDP Inc
Priority date: 1999-11-17
Filing date: 2000-11-16
Publication date: 2002-09-11
Also published as: WO2001037044A1; CN1391664A; WO2001037043A1; KR20020053083A; TW594378B; HK1052751A1; JP2003514955A; JP2003514956A; KR20030008215A; EP1238310A1

Abstract

Disclosed are partially fluorinated and fully fluorinated polymers that are substantially transparent to ultraviolet radiation at wavelengths from 140 to 186 nanometers.

Description

TITLE

ULTRAVIOLET AND VACUUM ULTRAVIOLET TRANSPARENT

POLYMER COMPOSITIONS AND THEIR USES

FIELD OF THE INVENTION This invention concerns partially fluorinated and fully fluorinated polymers that are substantially transparent to ultraviolet radiation at wavelengths from approximately 140 nanometer to 186 nanometers.

TECHNICAL BACKGROUND OF THE INVENTION The semiconductor industry is the foundation of the trillion dollar electronics industry. The semiconductor industry continues to meet the demands of Moore's law, whereby integrated circuit density doubles every 18 months, in large part because of continuous improvement of optical lithography's ability to print smaller features on silicon. The circuit pattern is contained in the photomask, and an optical stepper is used to project this mask pattern into the photoresist layer on the silicon wafer. Current lithography is done using 248 nm light; lithography with 193 nm light is just entering early production. Alternate methods of lithography that do not use visible or ultraviolet light waves, i.e., the next generation lithographies utilizing X-rays, e-beams or EUV radiation have not matured sufficiently that they are ready to be adopted for production. The industry, through its international consortium SEMATECH, has concluded that 157 nm light based lithography will be the next technological step. 157 nm lies in the region of the spectrum referred to as the vacuum uv (VUV), which range extends from 186 nm down to below 50 nm. Use of this VUV lithography requires materials transparent in this range. Specifically, the new 157 nm optical lithography standard leads to a demand for new polymeric materials, based on the requirement for transparency at 157 nm. As use of this new technology develops, there remains a continuing need for improved materials useful at the shorter wavelengths.

Certain fluoropolymers have already been identified in the art as useful for optical applications such as light guides, anti-reflective coatings and layers, pellicles, and glues. Most of this work has been done at wavelengths above 200 nm where perfluoropolymer absorption is of little concern.

WO 9836324, August 20, 1998, Mitsui Chemical Inc., discloses the use of resins consisting solely of C and F, optionally in combination with silicone polymers having siloxane backbones, as pellicle membranes having an absorbance/micrometer of 0.1 to 1.0 at UV wavelengths from 140 to 200 nm. Data in the literature, together with applicant's measurements, for fluoropolymers (see Table I below) demonstrate that, at least at 157 nm, C and F fluoropolymers have absorbances much larger than A/μ = 0.1 to 1 as claimed by WO 9836324. WO 9822851, May 28, 1998, Mitsui Chemicals, Inc., claims the use of photodegradation-resistant, tacky polymers that immobilize dust particles when coated on the inside of a pellicle frame. These tacky materials have compositions consisting largely of low molecular weight -(CF₂-CXR) copolymers in which X is halogen and R is -Cl or -CF3. Higher molecular weight polymers such as poly(perfluorobutenyl viny ether), poly[(tetrafluoroethylene/perfluoro- (2,2-dimethyl- 1 ,3-dioxole)], poly(tetrafluoroethylene/hexafluoropropylene/vinylidene fluoride), poly(hexafluoropropylene/vinylidene fluoride), or poly(chlorotolyl fluoroethylene/vinylidene fluoride) are added as a minor component to improve creep resistance. It should be noted that all exemplifications of this technology were with poly(chlorotrifluoroethylene) as the low molecular weight adhesive agent, that poly(chlorotrifluoroethylene) absorbs 157 nm light strongly, and that retention of tackiness after extended UV degradation (illustrated only at 248 nm), not transparency, was the only demonstrated advantage of the claimed formulations.

Japanese Patent 07295207, November 10, 1995, Shinetsu Chem. Ind Co, claims double layer pellicles combining Cytop™ CTXS

(poly(CF₂=CFOCF₂CF₂CF=CF₂)) with Teflon® AF 1600 for greater strength. Both Teflon® AF 1600 and Cytop™ absorb strongly at 157 nm (see Table 2). US Patent 5286567, February 15, 1994, Shin-Etsu Chemical Co., Ltd., claims the use of copolymers of tetrafluoroethylene and five membered cyclic perfluoroether monomers as pellicles once they have been made hydrophilic, and therefore antistatic, by plasma treatment. That the presence of five membered ring monomers and tetrafluoroethylene is not sufficient criteria to yield A/μ < 0.1, is illustrated in Table 2 where only one out of five such polymers meet this goal. European Patent 416528, March 13, 1991, DuPont, claims amorphous fluoropolymers having a refractive index of 1.24-1.41 as pellicles at wavelengths of 190-820 nm.

Japanese Patent 01241557, Bando Chemical Industries, Ltd., September 26, 1989, claims pellicles usable at 280-360 nm using (co)polymers of vinylidene fluoride (VF₂,), tetrafluoroethylene/hexafiuoropropylene (TFE/HFP), ethylene/tetrafluoroethylene (E/TFE), TFE/CF₂=CFORf, TFE/HFP/CF₂=CFORf, chlorotrifluoroethylene (CTFE), E/CTFE, CTFE/VF₂ and vinyl fluoride (VF). Japanese Patent 59048766, March 21, 1984, Mitsui Toatsu Chemicals, Inc., claims the use of a stretched film of poly(vinylidene fluoride) as having good transparency from 200 to 400 nm.

Many of the fluoropolymers cited in the references above are noticeably hazy to the eye because of crystallinity and would therefore be expected to scatter light to a degree unsuitable for high light transmission and the accurate reproduction of circuit patterns. Poly(vinylidene fluoride), poly(chlorotri- fluoroethylene), poly(tetrafluoroethylene/ethylene), commercially available poly(tetrafluoroethylene/hexafluoropropylene) compositions, and poly(ethylene/chlorotrifluoroethylene) are all such crystalline, optically hazy materials. More recent references have thus been directed at Cytop™ and Teflon® AF because they combine perfluorination with outstanding optical clarity, solubility, and a complete lack of crystallinity. As shown below in the Discussion, however, Cytop™ and most grades of Teflon® AF do not have the required transparency at 157 nm.

It is an object of the present invention to overcome the difficulties associated with the prior art by providing partially fluorinated and fully fluorinated polymers that are substantially transparent to ultraviolet radiation at wavelengths between 140 and 186 nanometers, especially at 157 nm. SUMMARY OF THE INVENTION

This invention provides a vacuum ultraviolet (VUV) transparent material exhibiting an absorbance/micron (A/micrometer) < 1 at wavelengths from 140 to 186 nanometers comprising amorphous vinyl homopolymers of perfluoro-2,2- dimethyl-1,3- dioxole or CX₂=CY₂ where X = F or CF3 and Y = -H or amorphous vinyl copolymers of perfluoro-2,2-dimethyl- 1 ,3-dioxole and CX₂=CY₂, said homopolymers or copolymers optionally containing 0 to 25 mole % of one or more monomers CR^aR^b=CR^cR^d in the case where CR^aR^b=CR^cR^d enters the homopolymer or copolymer in approximately random fashion and optionally 40 to 60 mole % of one or more monomers CR^aR^b=CR^cR^d in the case where the CR^^CR^*¹ enters the copolymer in approximately alternating fashion, where each of the R^a, R^b, and R^c is selected independently from H or F and where R^d is selected from the group consisting of -F, -CF3, -ORf where Rf is C_nF_2n+1 with n = 1 to 3, and -OH (when R^c =H).

This invention also provides a vacuum ultraviolet (VUV) transparent material exhibiting an absorbance/micron (A/micrometer) < 1 at wavelengths from 140-186 nm comprising amorphous vinyl copolymers of CH₂=CHCF3 and CF₂=CF₂; CH₂=CFH and CF₂=CFC1; CH₂=CHF and CClH=CF₂.wherein the ratio of monomers ranges from approximately 1:2 to approximately 2:1; perfluoro(2-methylene-4-methyl-l ,3-dioxolane) and perfluoro(2,2-dimethyl-l ,3- dioxole); perfluoro(2-methylene-4-methyl-l,3-dioxolane) and vinylidene fluoride in any ratio that gives an amorphous composition; and the homopolymer of perfluoro(2-methylene-4-methyl- 1 ,3-dioxolane).

This invention further provides pellicles, anti-reflective coatings, optically clear glues, light guides and resists comprising the UV transparent material provided above.

This invention further provides copolymer compositions comprising poly(hexafluoroisobutylene:trifluoroethylene) with 40-60 mole % hexafluoroisobutylene and 60-40 mole % trifluoroethylene and copolymer compositions comprising poly(hexafluoroisobutylene: vinyl fluoride) with 40-60 mole % hexafluoroisobutylene and 60-40 mole % vinyl fluoride. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 describes the 157 nm transmission T of a pellicle in units of % as a function of the 157 nm absorbance, in units of inverse microns, of the polymer for an absorbance range from 0.4 to 0.0. The effects of thin film interference in the pellicle membrane are neglected in this calculation.

Figure 2 describes the absorbance in units of inverse microns for Teflon® AF 1601 (Sample 7a) and Cytop™ (Sample 13) versus wavelength lambda (λ) in units of nanometers.

Figure 3 describes the index of refraction n and the extinction coefficient k determined for a Teflon® AF 1601 film (Sample 7b) of thickness 2146 angstroms on a silicon substrate versus wavelength lambda in units of nanometers determined by VUV spectroscopic ellipsometry.

Figure 4 describes the spectral transmission in absolute units versus the wavelength lambda in units of nanometers for a pellicle of Teflon® AF 1601 designed as an unsupported tuned etalon with a film thickness of 6059 angstroms. The interference fringes of the tuned etalon are clearly visible as a function of wavelength.

Figure 5 describes the spectral reflectance in absolute units versus the wavelength lambda in units of nanometers for a pellicle of Teflon® AF 1601 designed as an unsupported tuned etalon with a film thickness of 6059 angstroms. The interference fringes of the tuned etalon are clearly visible as a function of wavelength, and a minimum in the pellicle reflectance is seen at 157 nm which contributes to the maximized pellicle transmission at this lithographic wavelength. Figure 6 describes transmission of a tuned etalon pellicle film of Teflon® AF 1601 at a lithographic wavelength of 157 nm as a function of the pellicle film thickness. The oscillations in the pellicle transmission with thickness arise due the thin film interference fringes in the film and give rise to pellicle transmission maxima and minima. The optimum tuned etalon pellicle design will correspond to the film with sufficient mechanical integrity and a thickness such that the transmission is at a maxima. Still as can be seen, pellicles designed from this material have substantially lower transmissions than the target transmission for a 157 nm pellicle. Figure 7 describes the absorbance in units of inverse microns for

Teflon® AF 1200 (Sample 8), Teflon® AF 1601 (Sample 7a), and Teflon® AF 2400 (Sample 5)versus wavelength lambda (λ) in units of nanometers. Notice the dramatic decrease in the absorbance/micron as the PDD content of the polymer increases and the TFE content of the polymer decreases from 52% to 32% to 11%, and therefore the lengths of any (CF₂)_n runs in the polymer decreases.

Figure 8 describes the absorbance in units of inverse microns for TFE:HFP (Sample 14) and TrFE:HFP (Sample 12) versus wavelength lambda (λ) in units of nanometers. The presence of the HF carbons in the CF₂=CFH monomer interrupts extended CF₂ runs. This effect can also be understood as the absorption maxima of TFE:HFP polymer shifts to shorter wavelengths in the TrFE:HFP polymer.

Figure 9 describes the absorbance in units of inverse microns for VF₂:PDD (Sample 2), VF₂:HFP (Sample 1), HFIB:TrFE (Sample 3) and HFIB:VF (Sample 4) versus wavelength lambda (λ) in units of nanometers. Figure 10 describes the index of refraction n and the extinction coefficient k determined for a HFIB:VF film (Sample 4a) of thickness 14,386 angstroms on a silicon substrate versus wavelength lambda in units of nanometers determined by VUV spectroscopic ellipsometry.

Figure 11 describes the spectral transmission in absolute units versus the wavelength lambda in units of nanometers for a pellicle of HFIB: VF designed as an unsupported tuned etalon with a film thickness of 3660 angstroms. The interference fringes of the tuned etalon are clearly visible as a function of wavelength.

Figure 12 describes the spectral reflectance in absolute units versus the wavelength lambda in units of nanometers for a pellicle of HFIB : VF designed as an unsupported tuned etalon with a film thickness of 3660 angstroms. The interference fringes of the tuned etalon are clearly visible as a function of wavelength, and a minimum in the pellicle reflectance is seen at 157 nm which contributes to the maximized pellicle transmission at this lithographic wavelength.

Figure 13 describes transmission of a tuned etalon pellicle film of HFIB:VF with an absorbance per micron of 0.022 and an index of refraction of 1.5 at a lithographic wavelength of 157 nm as a function of the pellicle film thickness. Note that for pellicle film thicknesses up to 3660 angstroms, the maximum pellicle transmission is above the target specification of 98%.

Figure 14 describes transmission of a tuned etalon pellicle film of a polymer with an absorbance per micron of 0.01 and an index of refraction of 1.5 at a lithographic wavelength of 157 nm as a function of the pellicle film thickness. Note that for pellicle film thicknesses up to 8371 angstroms, the maximum pellicle transmission is above the target specification of 98 % .

Figure 15 describes the absorbance in units of inverse microns for 5:6 TFP:TFE (Sample 17), HFIB:VF (Sample 18), 5:2 VF2:PFMVE (Sample 19), 7:5 VF2:PFPVE (Sample 21) and 79:21 VF2:HFP (Sample 22) versus wavelength lambda (λ) in units of nanometers.

Figure 16 describes the absorbance in units of inverse microns for 1:1 PDD:TrFE (Sample 9), 13:10 VF2:PFMVE (Sample 20), 2:5:2 HFP:PFMVE:VF2 (Sample 23), 10:7 HFIB:VF (Sample 24) and 6:5 PDD:PFMVE (Sample 25) versus wavelength lambda (λ) in units of nanometers.

Figure 17 describes the absorbance in units of inverse microns for 20:11 VF:C1DFE (Sample 26), 1 :2 PDD:VF2 (Sample 27), 1:1 HFIB:VA (Sample 32), PMD (Sample 33) and PMD:PDD (Solution 34) versus wavelength lambda (λ) in units of nanometers. Figure 18 describes the absorbance in units of inverse microns for 1 :1

CTFE:VF (Sample 6b), 5:2 VF2:TrFE (Sample 10), 10:23 PDD:CTFE (Sample 11), 5:4 VF2:CTFE (Sample 15) and 1:1 PMD:TFE (Sample 18) versus wavelength lambda (λ) in units of nanometers.

Figure 19 describes the absorbance in units of inverse microns for 5:8 PDD:VF2 (Sample 29) and 41:37:22 HFIB:VF:VF2 (Sample 19) versus wavelength lambda (λ) in units of nanometers.

DETAILED DESCRIPTION OF THE INVENTION The invention provides fluoropolymer compositions and their use in certain electronic applications. Certain abbreviations used throughout the specification are tabulated: TFE tetrafluoroethylene

HFP hexafluoropropylene

VF vinylfluoride CTFE chlorotrifluoroethylene

VF vinylidene fluoride

HFIB hexafluoroisobutylene

TrFE trifluoroethylene

PDD 4,5-difluoro-2,2-bis(trifluoromethyl)-l ,3- dioxole

PMD Perfluro(2-methylene-4-methyl- 1 ,3- dioxolane)

Teflon® AF 2400 89:11 PDD:TFE

Teflon® AF 1601 68:32 PDD:TFE

Teflon® AF 1200 48:52 PDD:TFE

C1DFE 1 -chloro-2,2-difluoroethylene

PPVE perfluoro(propyl vinyl ether)

PMVE perfluoro(methyl vinyl ether)

VOAc vinyl acetate V VOOHH vinyl alcohol

TrP 3 ,3 ,3 -trifluoropropene

Fluorinert® FC-75 Electronic fluid manufactured by 3M, believed to appoximate perfluoro(butyl tetrahydrofuran)

Fluorinert® FC-40 Electronic fluid manufactured by 3M, believed to approximate perfluoro(tributylamine)

Vazo® 56 WSP Initiator manufactured by DuPont, 2,2'- bis(2-amidino-propane)dihydrochloride

DSC Differential scanning calorimetry

Use of 157 nm light in the next generation of lithography raises new photochemical issues not encountered in lithography at longer wavelengths. The energy of a 157 nm photon (182 kcal/mole or 7.9 eV of light quanta) is great enough to break common chemical bonds as, for example, C-F (108-116 kcal/mole), C-H (98-105 kcal/mole), C-C (88-97 kcal/mole), and C-Cl (82-86 kcal/mole) bonds. Absorption maxima for selected hydrocarbon and fluorocarbon compounds are shown in Table 1. TABLE 1 Comparison of UV Absorption Maxima for Hydrocarbons and Fluorocarbons

^lB. A. Lombos, P. Sauvageau, and C. Sandorfy, Chem. Phys. Lett.. 1967, 42. ²K. Seki, H. Tanaka, T. Ohta, Y. Aoki, A, Imamura, H. Fujimoto, H. Yamamoto, H. Inokuchi. Phvs. Scripta. 41, 167(1990).

As can be seen from the table, UV absorption maxima move to longer wavelengths as chain length increases for both hydrocarbons and fluorocarbons. Fluorocarbon chains (CF₂)n absorb at 157 nm somewhere between n = 6 (142 nm) and n = 172 (161 nm) while hydrocarbon chains (CH₂)n absorb at 157 nm as early as n = 2. Clearly then fluorocarbon chains are more resistant than hydrocarbon chains to UV absorption and not surprisingly the industry has been moving increasingly towards perfluorination when seeking high UV transmission. But, as long as chain lengths still offering acceptable transparency can never exceed (CH₂) i or (CF₂)G, perfectly transparent polymers at 157 nm would seem precluded. Consistent with this, V. N. Vasilets, et al., J Poly. Sci, Part A, Poly. Chem., 36, 2215(1998) for example report that poly(tetrafluoro- ethylene/hexafluoropropylene), [poly(TFE/HFP)], (Teflon® AF FEP), shows strong absorption and photochemical degradation at 147 nm. Similarly we find that 1 : 1 poly(hexafluoropropylene:tetrafluoroethylene) is highly absorbing at 157 nm (Table 2, A/micron - 3.6 @ 157 nm).

Polymers play a critical role in lithography in multiple areas: one is the polymer pellicle which is placed over the mask pattern to keep any particulate contaminants out of the photomask object plane, thereby ensuring that the lithographic imaging will be defect free. The pellicle is a free standing polymer membrane, typically 0.8 micrometers in thickness, which is mounted on a typically 5 inch square frame. The pellicle film must have high transparency or transmission of light at the lithographic wavelength for efficient image formation and must neither darken nor burst with prolonged illumination in the optical stepper. Pellicles for current lithographic wavelengths utilize pellicles with > 99% transmission, through exploitation of polymers with very low optical absorption combined with thin film interference effects. SEMATECH's goal for pellicles for use in 157 nm photolithography is greater than 98% transparency over an exposure lifetime of 75 million laser pulses of 0.1 mJ/cm², or a radiation dose of 7.5 kJ of 157 nm light. A pellicle transmission of 98% corresponds to an absorbance A of approximately 0.01 per micrometer of film thickness. The absorbance is defined in Equation 1, where the Absorbance A per micron of film thickness, is defined as the base 10 logarithm of the ratio of the substrate transmission divided by the transmission of the sample consisting of the polymer film sample on its substrate, this quantity divided by the polymer film thickness.

Equation 1. j , -1 _{Λ Λ I} tfitm

In this manner the Absorbance A has units of inverse microns (or 1 /micron, where a micron, is a micrometer or um of polymer film thickness. The absorbance/micron of polymer films discussed here were measured for polymer films spun coated on to CaF₂ substrates using standard methods. The VUV transmission of each CaF₂ substrate was measured prior to the spin coating of the polymer film. Then the VUV transmission of the polymer film on that particular CaF₂ substrate was measured, and using the measured film thickness (reported in Table 2) and equation 1, the values of the absorbance/micron for the polymers, as a function of wavelength was determined, and the value of the absorbance/micron for a wavelength of 157 nm is tabulated in Table 2. For some materials, two films, of differing thicknesses are presented in the Table 2, and Absorbance/micron values for each film are also presented.

The VUV transmission of the CaF₂ substrates and the polymer films on the CaF₂ substrates were measured using a VUV spectrophotometer using a laser plasma light source, a sample chamber capable of both transmission and reflectance measurements, a 1 meter monochromator and a sodium salicylate phosphor coated 1024 element photodiode detector. This is discussed in greater detail in R. H. French, "Laser-Plasma Sourced, Temperature Dependent VUV Spectrophotometer Using Dispersive Analysis", Physica Scripta, 41, 4, 404-8, (1990) which is incorporated herein by reference.

The absorbance per micron of a polymer will determine the average transmission of an unsupported pellicle film made from that polymer. In Figure 1 , the 157 nm transmission T of a pellicle in units of % as a function of the 157 nm absorbance, in units of inverse microns, of the polymer for an absorbance range from 0.4 to 0.0 is shown. The effects of thin film interference in the pellicle membrane are neglected in this calculation. The results for pellicle films of thicknesses ranging from 0.2 microns to 1 micron are shown, and demonstrate that for any particular polymer, the pellicle transmission can be increased, through the use of a thinner pellicle film thickness. This approach to increasing the pellicle transmission has a limited range of utility, since the pellicle film is an unsupported polymer membrane and must have sufficient mechanical strength and integrity. These mechanical requirements suggest the use of polymer with relatively high glass transition temperature T_g and polymer film thicknesses of 0.6 microns or greater. From Figure 1 it can be seen that a target absorbance per micron of a polymer for pellicle applications is < 0.02 abs./micron at 157 nm.

In Figure 2 the absorbance in units of inverse microns for Teflon® AF 1601 and Cytop™ versus wavelength lambda (λ) in units of nanometers is shown. At 157 nanometers, the absorbance/micron for Cytop has a value of 1.9/micron while Teflon® AF 1601 has an absorbance of 0.42/micron, approximately 5 times lower than Cytop. In Example 1 we will demonstrate that even the 157 nm absorbance of Teflon® AF 1601 of 0.42/micron is too high for application as a pellicle for use at 157 nm. The optical properties (index of refraction, "n" and extinction coefficient,

"k") are determined from variable angle spectroscopic ellipsometry (VASE) at three incident angles covering the wavelength range from 186-800 nm, corresponding to an energy range of 1.5-6.65 eV, in combination with VUV ellipsometry (VUV-VASE) measurements performed at a single angle of incidence from 143-275 nm, corresponding to an energy range of 4.5-8.67 eV. The polymer films were spin coated onto a silicon substrate. The VASE ellipsometers were manufactured by J. A. Woollam Company, 645 M Street, Suite 102, Lincoln, NE 68508 USA. Optical constants were fit to these data simultaneously, using an optical model of the film on the substrate. See generally, O. S. Heavens, Optical Properties of Thin Solid Films, pp. 55-62, Dover, NY, 1991, incorporated herein by reference. From knowledge of the spectral dependence of optical properties, the transmission of a pellicle film of arbitrary thickness can be calculated by using an optical model for the unsupported pellicle film, at a specific polymer film thickness, and then calculating the pellicle film transmission and reflectance. In this manner the pellicle film thickness can be optimized such that the pellicle will exhibit a thin film interference maximum in the transmission spectrum at the desired lithographic wavelength. These transmission maxima occur for various film thicknesses, determined from the index of refraction of the polymer, the lithographic wavelength of interest and the film thickness. The transmission maxima of a properly tuned etalon pellicle film occur where the reflectance of the pellicle film exhibits minimum in the reflectance, and correspond to minimizing the pellicle's reflectance and maximizing its transmission at the lithographic wavelength. The relationship between the extinction coefficient k and the absorption coefficient α and the absorbance per micron A is given in Equation 2, where lambda is the wavelength of light. This relationship is useful in comparing results of absorbance measurements and ellipsometry measurements. This relationship for A is exact if light scattering in the polymer film (as may occur due to crystallinity of the polymer), thin film interference effects, and surface scattering effects are minimized.

Equation 2.

aλ ~ ~ Aλ k = — = 2.3

4π Aπ

Polymeric materials with very low absorbance/micron or extinction coefficients and low values of the index of refraction also have very important applications as anti-reflection coatings and optical adhesives. A low absorbance material, as taught here in can be used to reduce the light reflected from the surface of a transparent substrate of a relatively higher index of refraction. This decrease in the reflected light, leads to a concomitant increase in the light transmitted through the transparent substrate material. This anti reflective coating effect of these low absorbance/micron materials can be seen in the results for VF₂:PDD, VF₂:HFP, HFIB:TrFE, and HFIB:VF where the absorbance of the polymers on CaF₂ substrates for the case of very thin films, exhibits negative absorbance/micron. This corresponds to an increase of 157 nm light transmission through the polymer film on the CaF₂ substrate as compared to the light transmission through the bare CaF₂ substrate. For VF₂:HFP, HFIB:TrFE and HFIB:VF we have measured much thicker polymer films (also listed in Table 2) in which this antireflective coating effect is not seen, where the absorbance/micron of the polymers is seen to be positive and very small.

Polymers such as these can also be used as adhesives to join optical elements together, and since they have low optical absorbance/micron and low values of the index of refraction, they serve to reduce the reflectance of light at the air/substrate interfaces among the optical elements, and serve to direct more of the transmitted light from one optical element into subsequent optical elements in the system. The materials of the present invention are useful in the manufacture of transmissive optical elements, such as lenses and beam splitters, for use in the vacuum UV region.

These materials may also be used as elements in a compound lens designed to reduce chromatic aberrations. At present only CaF₂ and possibly hydroxyl free silica are viewed as having sufficient transparency at 157 nm to be used in transmissive focussing elements. It is also commonly known (e.g, see R. Kingslake, Academic Press, Inc., 1978, Lens Design Fundamentals, p. 77) that by using a second material of different refractive index and dispersion, an achromatic lens can be created. A Sellmeier fit to the data shown in Figure 10 for HFIB:VF as described in Example 4 shows it to have a refractive index of 1.4942 and a dispersion of 0.00220 nm^{- 1} at 157 nm. A similar fit to the index of refraction data for CaF₂ from Edward D. Palik, Handbook of Optical Constants of Solids II, p. 831, Academic Press, Inc., Boston, MA (1991) and French shows it to have an index 1.5584 and a dispersion of 0.00234 nm"¹ at 157 nm. Thus, by using one of these materials in conjunction with CaF₂, it is expected that an achromatic lens can be constructed from this and other similar materials described in this application.

An additional area in which polymers play a critical role is as the photosensitive photoresist which captures the optical latent image. In the case of photoresists, light must penetrate the full thickness of the resist layer for a latent optical image, with well defined vertical side walls to be produced during optical imaging which then will produce the desired resist image in the developed polymer. When used as a resist at 157 nm, a polymer can have a considerably higher absorption coefficient of A < -2-3 per micrometer of film thickness, if the resist thickness is limited to about 2000 A.

WO 9836324 discloses carbon/fluorine polymers such as poly(tetrafluoroethylene/hexafluoropropylene) as pellicle membranes having absorption A/μ of 0.1 to 1 for use at 140 to 200 nm. The data in Table 1 above, derived from literature sources, Vasilets' report of high absorption and photodegradation of poly(hexafluoropropylene/tetrafluoroethylene) at 148 nm [V. N. Vasilets, et al, J Poly. Set, Part A, Poly. Chem., 36, 2215 (1998), and the data in Table 2 which combines additional literature data with applicant's data, casts doubt on this disclosure. For example, poly(tetrafluoroethylene:hexafluoropropylene), polymer #14 of Table 2, shows a relatively strong A/μ of 3.9 at 157 nm that fails not only the industry goals for pellicles (A/μ < 0.01) but also the much looser goals for resists (A/μ < 2-3). Japanese Patent 072952076 claims bilayer membranes of Cytop™ and Teflon® AF 1600 as pellicle films. At 157 nm Cytop™ has an A/μ of 1.9 (polymer #13, Table 2) and Teflon® AF 1600 an A/μ of 0.4 (polymer #7, Table 2). This is not surprising considering the data of Table 1 that shows significant UV light absorption at 160 nm whenever there are more than about 6 CF₂ groups connected in a chain. Indeed, W. H. Buck and P. R. Resnick report that Teflon(TM) AF 1600 has more than 30% of its CF₂ units present as (CF₂)n runs with n >6 consistent with its A/u =0.4 [J. Scheirs, Modern Fluoropolymers, John Wiley, New York, 1997, Chapter 22, page 401]. Unless the interactions responsible for such absorption can be broken up, it would seem impossible to find a carbon based polymer completely transparent at wavelengths shorter than about 160 nm.

Table 2 below lists absorbance/micrometer (A μ) at 157 nm for partially and fully fluorinated polymer films that have been spin coated on CaF₂ crystals. Polymers are listed in order of increasing absorption. In some cases, more than one sample of various polymers have been prepared within an Example.

Therefore, Table entries are identified by both Example number (first column) and sample number. In addition, reference to Figures displaying spectra of various polymers are cross referenced in the Table.

TABLE 2

Absorbance per micron (A/μm) for Selected Fluoropolymers

t

CO

to ID

o

(A/μm) determined from VUV transmission based absorbance measurements,one result, marked by ^# was determined from VUV ellipsometry

OO

'J. Scheirs, editor, Modern Fluoroplastics, John Wiley U Sons, West Sussex, England, 1997, Chapter 28

²PDD = 4,5-difluoro-2,2-bis(trifluoromethyl)-l,3-dioxole

3pMD = Perfluro(2-methylene-4-methyl-l,3-dioxolane

⁴Teflon® AF 2400

⁵Teflon® AF 1601

⁶Teflon® AF 1200

^Monomers were loaded at 3:4 mole ratio. Polymer product not analyzed

Three of the acyclic polymer structures listed in Table 2 show no detectable UV absorption at 157 nm (polymers VF₂:HFP, HFIB:TrFE, and HFIB:VF). These are copolymers containing either vinylidene fluoride (VF₂) or hexafluoroisobutylene (HFIB). VF₂ and HFIB have a unique structural feature in common. Taken by itself, for example, the hexafluoroisobutylene monomer can not form CH₂ runs longer than CH₂CH before the run is broken up by a -C(CF3) - segment. Similarly vinylidene fluoride can not form of CH₂ runs longer than CH CH before they are broken up by a CF or CF₂ runs longer than CF₂CF₂ before they broken up by a CH . That is, CX₂=CY₂ monomers such as VF₂ and HFIB have the property of being "self-interrupting" when judging the potential for extended interactions between C-F bonds or extended interactions between C-H bonds. The stiff five membered ring of PDD likely has a related effect by forcing conformations unfavorable for interaction. Comparing PDD/TFE copolymers #5, #7, and #8, absorption falls off from 0.6 to 0.4 to 0.0 as TFE content drops from 52 to 32 to 11 mole %. That is, transparency improves as increasing PDD content interrupts (CF₂)_n runs. It is anticipated that other perfluorinated ring structures or partially fluorinated ring structures can be found that will serve the same interrupting function as PDD.

Solutions of TFE:HFP and TrFE:HFP were spin coated at spin speeds of 6000 rpm onto CaF substrates to produce polymer films of 1850 angstroms and 1389 angstroms thicknesses respectively. VUV absorbance measurements were then used to determine the absorbance per micron.

The introduction of TrFE in place of TFE demonstrates another example of decreasing the 157 nm absorbance/micron. The absorbance in units of inverse microns for TFE:HFP and TrFE:HFP versus wavelength lambda (λ) in units of nanometers is shown in Figure 8. The presence of the CHF carbons in the CF₂=CFH monomer interrupts extended CF runs.

This produces a decrease in the 157 nm absorbance of TFE:HFP of 3.9/micron to an absorbance/micron of 1.37/micron for TrFE/HFP. This effect can also be understood as the absorption maxima of TFE:HFP polymer shifts to shorter wavelengths in the TrFE:HFP polymer.

Thus, we define a class of polymers highly transparent (A/μ < 1, more preferably A/u < 0.1) to light of < 186 nm made by homo or copolymerizing PDD and CX =CY monomers (X = -F or -CF₃ and Y = -H) and optionally other monomers CR^aR^b=CR^cR^d where any one or all of R^a, R^b or R^c, may be H or F and where R^d may be F, -CF₃, -ORf where-Rf is C_nF_2n+ι with n = 1 to 3, and OH (when R^c =H) as needed to introduce solubility or break crystallinity. When the CRaRb=CRcRd monomer polymerizes randomly with PDD or the CX₂=C Y₂ monomer, it can not be present to greater than -25 mole percent because at higher concentrations the statistics start to allow CR^aR^b=CR^cR^d runs long enough for noticeable absorption. As mentioned above, this is illustrated by the PDD:TFE copolymers in which A/μ drops from 0.6 to 0.4 to 0.0 as TFE content decreases from ~52 to 32 to 11 mole percent (Table 2, polymers Teflon® AF 1200,

Teflon® AF 1601, Teflon® AF 2400 respectively). When the CR^aR^b=CR^cR^d monomer polymerizes in an alternating fashion with PDD or the CX₂=CY₂ monomer, a higher concentration of the CR^aR^b=CR^cR^d monomer can be tolerated. An example of this is the 1 : 1 HFIB:TrFE copolymer which had an A/μ of -0.05 (Table 2, polymer #3). Random and alternating structures of course are never 100% ideal and some monomers naturally tend to either block or avoid polymerizing with themselves so the limits of 25% CR R^b=CR^cR^d for random structures and 50% CR^aR^b=CR^cR^d for alternating structures are approximate. Other highly transparent combinations include ~2:1 to 1:2 copolymers of CH₂=CHCF₃:CF₂=CF₂, CH₂=CHF:CF₂=CFC1, CH₂=CHF:CC1H=CF₂. Preferred monomers for CX₂=CY₂ include vinylidene fluoride and hexafluoroisobutylene. Preferred CR^aR^b=CR^cR^d monomers are ones that introduce an asymmetric center in the polymer chain such as vinyl fluoride, trifluoroethylene, hexafluoropropylene, and chlorotrifluoroethylene so as to increase solubility and break crystallinity. Finally it should be pointed out that making highly transparent structures as defined above does not automatically make a polymer useful for all claimed applications. It has already been pointed out for example that the crystallinity of poly( vinylidene fluoride) precludes its use as a perfectly clear and transparent optical material [that is the failure of poly(vinylidene fluoride) as a pellicle would not be in the absorption but for a physical reason (light scattering)]. As another example, poly(perfluorodimethyldixolole) is too insoluble to easily spin coat as thick films. In the case of poly(perfluorodimethyldioxole) this can be gotten around by coating liquid monomer and polymerizing in place. Films of thickness greater than about 250 nm can be prepared by placing monomer optionally diluted with solvent and/or initiator in the location where the film is desired . Polymerization can be initiated by appropriate physical and/or chemical means which leads to the deposition of polymer, as formed, in the desired location. The result, after subsequent solvent removal, is a film of polymer thicker than can be prepared by customary solvent coating techniques. As a final example, poly [vinylidene fluoride/perfluoro(methyl vinyl ether)] is a tacky gum useful for glues but not useful as a self-supporting pellicle film. As used herein, the term amorphous fluoropolymer means a fluoropolymer that exhibits no melting point when analyzed by Differential Scanning Calorimetry. No melting point means no melting associated thermal event of greater than 1 Joule/gram.

Listing a monomer as a precursor to transparent polymers is not meant to imply that it will either homopolymerize or form a copolymer with any other listed monomer. Hexafluoroisobutylene for example, does not form useful quantities of decent molecular weight homopolymer or copolymerize with tetrafluoroethylene under ordinary conditions. While these materials are being claimed for use at 140 to 186 nm, they also make excellent clear polymers at longer wavelengths, up to 800 nm, and may also be suitable for some applications at still shorter wavelengths. Specifically preferred wavelengths are 140-160 nm, 150-160 nm, more preferred and 155-159 most preferred.

EXAMPLES Many of the polymers tested were obtained as commercial samples. Teflon® AF 1200: DuPont, Wilmington, DE Teflon® AF 1600: DuPont Teflon® AF 2400: DuPont

Cytop™: Asahi Glass, poly[perfluoro(butenyl vinyl ether)] -1 :1 Poly(hexafluoropropylene:tetrafluoroethylene) was made by the procedure described in US Patent 5,478,905, December 26, 1995. Many of remaining polymer compositions are known to the art and were prepared by standard methods. Typically autoclaves were loaded with monomer, solvent and initiator and heated to start polymerization. Most polymerizations were initiated using hexafluoropropylene oxide dimer peroxide 1 (DP) at ambient temperatures. CF₃CF₂CF₂OCF(CF₃)(C=0)OO(C=O)CF(CF₃)OCF₂CF₂CF₃

1, DP DP was prepared and used as a 0.05 to 0.2 molar solution in solvents such as Vertrel™ XF (CF₃CFHCFHCF₂CF₃) or 3M's Performance Fluid PF-5080 (largely perfluorooctane). DP can be made conveniently by either a routine laboratory procedure [Chengue, et al, J. Org. Chem., 47, 2009 (1982)] or on demand by a jet mixer process (U.S. 5,962,746 of October 5, 1999). The reaction mixture was then recovered and the polymer isolated by evaporation or filtration. Polymer compositions were determined by elemental analysis or by NMR. Procedures are given for the preparation of new compositions of matter, and representative procedures are given for previously known compositions.

The polymer films were prepared by spin coating of the polymer solutions onto CaF₂ substrates and then the polymer film on substrate samples were subjected to a post apply bake so as to assure that no residual solvents were remaining in the polymer film. The post apply bake temperature was in the range from 120°C to 250°C on a hot plate for two to five minutes or in a vacuum oven overnight. The spin speeds for the samples are listed in Table 2.

Polymer film thicknesses were determined using single or multiple wavelength ellipsometry as discussed below, or by analysis using a Filmetrics Model F20 thin film measurement system (Filmetrics, Inc. 7675 Dagget St., Suite 140, San Diego, CA 92111-2255). Film Thicknesses are reported in Table 2. For some materials, two films, of differing thicknesses are presented in the table, and Absorbance/micron values for each film are presented. COMPARATIVE EXAMPLE A -Teflon® AF 1601

Solutions of Teflon^® AF 1601 were spin coated at spin speeds of 6000 rpm onto CaF₂ substrates to produce polymer films of 3323 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. Teflon® AF 1601 is a 68:32 PDD:TFE polymer which is currently used as a polymer for pellicles designed for use at lithographic wavelengths of 248 nm and 193 nm. The absorbance/micron for 157 nm light is 0.42/microns as determined from VUV absorbance measurements. From Figure 1 this corresponds to pellicle transmissions below 70% for pellicle films of only 0.2 microns thickness, of course these results do not consider the additional effects of the thin film interference effects which arise from a properly designed pellicle film.

To determine the VUV optical properties of Teflon® AF 1601, VUV ellipsometry was performed on a Teflon® AF 1601 polymer sample on a silicon wafer, and the index of refraction and extinction coefficient shown in Figure 3 was determined. The 157 nm index of refraction for Teflon® AF 1601 is 1.4251. The 157 nm extinction coefficient determined corresponds to an absorbance/micron of 0.35/micron and is also listed in Table 2.

With these Teflon® AF 1601 optical properties, and the methods of O. S. Heavens discussed above, one can design the tuned etalon pellicle film, whereby the reflectance of the unsupported pellicle film is minimized and the pellicle transmission is maximized. (An etalon is a thin film in which thin film interference effects such as constructive and destructive interference of the light from the front and back surfaces of the constant thickness film gives rise to optical fringes in the wavelength dependence of the reflectance or transmission of the thin film (Principles of Optics, a book by Max Born and Emil Wolf, Pergamon Press, New York, 6th Edition, copyright 1980, PP 329-333). For a pellicle film thickness of 6059 angstroms, a 157 nm pellicle transmission is 65.7% while the 157 nm pellicle reflectance is 0.4%. The spectral transmission in absolute units versus the wavelength lambda in units of nanometers for a pellicle of Teflon® AF 1601 designed as an unsupported tuned etalon with a film thickness of 6059 angstroms is shown in Figure 4. The interference fringes of the tuned etalon are clearly visible as a function of wavelength. The spectral reflectance in absolute units versus the wavelength lambda in units of nanometers for the pellicle of Teflon® AF 1601 designed as an unsupported tuned etalon with a film thickness of 6059 angstroms is shown in Figure 5. The interference fringes of the tuned etalon are clearly visible as a function of wavelength, and a minimum in the pellicle reflectance is seen at 157 nm which contributes to the maximized pellicle transmission at this lithographic wavelength. For a 157 nm pellicle film of Teflon® AF 1601 with a film thickness of 6335 angstroms, the tuned etalon will not be optimized for maximum pellicle transmission and the 157 nm pellicle transmission will be 59.4% while the 157 nm pellicle reflectance increases to 8.1%.

Transmission of a tuned etalon pellicle film of Teflon® AF 1601 at a lithographic wavelength of 157 nm as a function of the pellicle film thickness is shown in Figure 6. The oscillations in the pellicle transmission with thickness arise due the thin film interference fringes in the film and give rise to pellicle transmission maxima and minima. The optimum tuned etalon pellicle design will correspond to the film with sufficient mechanical integrity and a thickness such that the transmission is at a maxima. Still as can be seen, pellicles designed from this material have substantially lower transmissions than the target 98% transmission for a 157 nm pellicle. Pellicles designed from Teflon® AF 1601 are not able to achieve pellicle transmissions above 98%. Pellicles designed from Cytop™, which has a much higher 157 nm absorbance/micron, will have even lower 157 nm pellicle transmissions. This demonstrates that methods are needed to produce polymers with dramatically lower 157 nm absorbance/micron so as to meet the desired 157 nm transmission of 98% for a pellicle film. Therefore we need polymers with substantially lower absorbance/micron.

EXAMPLE 1 - PDD/TFE Solutions of Teflon® AF 1200, 1601, and 2400 were spin coated at spin speeds of 6000 rpm onto CaF₂ substrates to produce polymer films of 4066 angstroms, 3323 angstroms and 2133 angstroms thicknesses respectively. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for Teflon® AF 1200 versus wavelength lambda (λ) in units of nanometers is shown in Figure 7 for sample 8. The 157 nm absorbance/micron determined is 0.64/micron. The 193 nm absorbance/micron determined is 0.004/micron. The 248 nm absorbance/micron determined is -0.001/micron.

The absorbance in units of inverse microns for Teflon® AF 1601 versus wavelength lambda (λ) in units of nanometers is shown in Figure 7 for Sample 7a. The 157 nm absorbance/micron determined is 0.42/micron. The 193 nm absorbance/micron determined is 0.02/micron. The 248 nm absorbance/micron determined is 0.01 /micron.

The absorbance in units of inverse microns for Teflon® AF 2400 versus wavelength lambda (λ) in units of nanometers is shown in Figure 7 for Sample 5. The 157 nm absorbance/micron determined is 0.007/micron. The 193 nm absorbance/micron determined is -0.06/micron. The 248 nm absorbance/micron determined is -0.06/micron.

One way to dramatically reduce the 157 nm absorbance of a PDD: TFE polymer is by increasing the percentage of PDD in the polymer. The stiff five membered ring of PDD likely has a similar effect, as VF₂ and HFIB, of being "self-interrupting" when judging the potential for extended interactions between C-F bonds by forcing unfavorable conformations. This is shown in Figure 7 where the absorbance in units of inverse microns for Teflon® AF 1200, Teflon® AF 1601, and Teflon® AF 2400 versus wavelength lambda (λ) in units of nanometers demonstrate the decreasing 157 nm absorbance/micron for these polymers. Comparing PDD/TFE copolymers Teflon® AF 1200, 1601, and 2400, absorbance/micron at 157nm falls off from 0.6 to 0.4/micron to 0.01/micron as TFE content drops from 52 to 32 to 11 mole %. That is, transparency improves as increasing PDD content interrupts (CF₂)_n runs. It is anticipated that other perfluorinated ring structures or partially fluorinated ring structures can be found that will serve the same interrupting function as PDD. EXAMPLE 2 - HFIB/TrFE

Preparation of 1:1 Poly(hexafluoroisobutylene:Trifluoroethylene) A 75 ml stainless steel autoclave chilled to <-20°C was loaded with 25 ml of CC1₂FCF₂C1 and 5 ml of -0.17 M DP in Vertrel™ XF. The autoclave was cooled, evacuated, and further loaded with 10 g of trifluoroethylene and 20 g of hexafluoroisobutylene. The reaction mixture was shaken overnight at ambient temperatures (28 to 33°C), removed from the autoclave, and evaporated down to a glassy film that was further dried for 72 hours in a 75 °C vacuum oven. There was obtained 3.33 g of white solid. A solution was made by rolling 3.24 g of this polymer with 6.99 g of l-methoxy-2-propanol acetate (PGMEA). This solution was mixed with 0.2 g of chromatographic silica gel and 0.2 g of decolorizing carbon and then passed first through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™) and then through a 0.45 μ PTFE membrane syringe filter (Whatman, Autovial™). Fluorine NMR of this solution, integrating trifluoroethylene's CHF fluorine at - -180 to -220 ppm versus hexafluoroisobutylene' s CF₃ fluorines at - -58 to -65 ppm, found -1:1 hexafluoroisobutylene:trifluoroethylene. This solution was used to spin coat thick films on optical substrate for absorption measurements. Sample preparation and results:

Solutions of HFIB:TrFE were spin coated at spin speeds of 3000 rpm and 6000 rpm onto CaF₂ substrates to produce polymer films of 12,146 angstroms and 1500 angstroms thicknesses respectively. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 3 :2 HFIB:TrFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 9 for Sample 3a. The 157 nm absorbance/micron determined from the thicker polymer film is 0.012/micron. The 193 nm absorbance/micron determined is 0.005/micron. The 248 nm absorbance/micron determined is -0.001/micron. EXAMPLE 3 - HFIB/VF

Preparation of 3:2 Poly(hexafluoroisobutylene: Vinyl Fluoride) A 75 ml stainless steel autoclave chilled to < -20°C was loaded with 25 ml of CC1₂FCF₂C1 and 10 ml of -0.07 M DP in Vertrel™ XF. The autoclave was cooled, evacuated, and further loaded with 16 g of hexafluoroisobutylene and 5 g of vinyl fluoride. The reaction mixture was shaken overnight at ambient temperatures (18 to 26°C), removed as a thick gel from the autoclave, evaporated, pumped down further with a vacuum pump for 96 hours, and then dried 22 hours in a 75°C vacuum oven. There was obtained 19 g of friable white solid with a T_g of 58°C (nitrogen, 10°C/min, second heat) and no detectable T_m.. Calc for (HFIB)₃(VF)₂: 32.89% C 2.07% H

Found: 32.83% C 2.08% H

Solution preparation and results:

A solution was made by rolling 6 g of this polymer with 12 g of 2-heptanone. This solution was passed through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™) and the filtrate used to spin coat thick films on optical substrate for absorption measurements. Sample preparation and results:

Solutions of HFIB: VF were spin coated at spin speeds of 3000 rpm and 6000 rpm (revolutions per minute) onto CaF₂ substrates to produce polymer films of 14,386 angstroms and 2870 angstroms thicknesses respectively. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 1 : 1 HFIB:VF versus wavelength lambda (λ) in units of nanometers is shown in Figure 9 for Sample 4a. The 157 nm absorbance/micron determined from the thicker polymer film is 0.027/micron. The 193 nm absorbance/micron determined is 0.020/micron. The 248 nm absorbance/micron determined is 0.008/micron.

To determine the VUV optical properties of HFIB: VF, VUV ellipsometry was performed on an HFIB:VF polymer sample on a silicon wafer, and the index of refraction and extinction coefficient shown in Figure 10 was determined. The 157 nm index of refraction for Teflon® AF 1601 is 1.50. The 157 nm extinction coefficient determined corresponds to an absorbance/micron of 0.022/micron and is also listed in Table 2.

With these HFIB:VF optical properties, and the methods of O. S. Heavens discussed above, one can design the tuned etalon pellicle film, whereby the reflectance of the unsupported pellicle film is minimized and the pellicle transmission is maximized. For a pellicle film thickness of 3660 angstroms, a 157 nm pellicle transmission is 98% while the 157 nm pellicle reflectance is 0.08%. The spectral transmission in absolute units versus the wavelength lambda in units of nanometers for a pellicle of HFIB: VF designed as an unsupported tuned etalon with a film thickness of 3660 angstroms is shown in Figure 11. The interference fringes of the tuned etalon are clearly visible as a function of wavelength. The spectral reflectance in absolute units versus the wavelength lambda in units of nanometers for the pellicle of HFIB: VF designed as an unsupported tuned etalon with a film thickness of 3660 angstroms is shown in Figure 12. The interference fringes of the tuned etalon are clearly visible as a function of wavelength, and a minimum in the pellicle reflectance is seen at 157 nm which contributes to the maximized pellicle transmission at this lithographic wavelength. For a 157 nm pellicle film of HFIB: VF with a film thickness of 3660 angstroms, the tuned etalon is a 157 nm pellicle with a 157 nm pellicle transmission of 98%. Transmission of a tuned etalon pellicle film of HFIB : VF at a lithographic wavelength of 157 nm as a function of the pellicle film thickness is shown in Figure 13. The oscillations in the pellicle transmission with thickness arise due the thin film interference fringes in the film and give rise to pellicle transmission maxima and minima.

EXAMPLE 4 - VF₂/PDD Sample preparation and results: A 75 ml autoclave was loaded with 25 ml of CF₂C1CC1₂F, prechilled to <-20°C, loaded further with 0.3 g of Perkadox(TM) 16N [bis(4-t-butylcyclohexyl)peroxydicarbonate] and 20 g of perfiuorodimethyldioxole (PDD), evacuated, and 10.4 g of vinylidene fluoride (VF₂) added. Heating 18 hours at 70°C gave a thick oil. Evaporation in air, drying 22 hours under pump vacuum, and drying 24 hours in a 75°C oven gave 17 g of hard white foam.

Elemental analysis, found: 28.99% C 1.11 % H

Calculated (C₅F₈O₂) 1 (C₂H₂F₂) 1 : 29.05%C 1.08% H

DSC, 10°C/min, nitrogen: Tg = 56°C (first heat)

Tg not detected second heat A viscous solution was prepared by rolling 4 g of poly(PDD VF₂) with 16 g of hexafluorobenzene and filtering through an 0.45 micron glass fiber syringe filter [Whatman, Autovial(TM)]. The filtrate was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of VF₂:PDD were spin coated at spin speeds of 6000 rpm onto CaF₂ substrates to produce polymer films of 2097 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for VF₂:PDD versus wavelength lambda (λ) in units of nanometers is shown in Figure 9 for Sample 2. The 157 nm absorbance/micron determined from the thicker polymer film is -0.04/micron. For this very transparent material, this demonstrates an antireflective coating effect thereby producing an negative value of the 157 nm absorbance/micron. The 193 nm absorbance/micron determined is 0.02/micron. The 248 nm absorbance/micron determined is 0.08/micron.

EXAMPLE 5 - VF₂/HFP Sample preparation and results: A poly(vinylidene fluoride/hexafluoropropylene) sample made by the method of US Patent 4,985,520 of 1/15/91) was characterized.

Composition by Fluorine NMR in Deuteroacetone: 21 mole % VF₂, 79 mole % HFP DSC, 10°C/min, nitrogen: Tg = -22.5°C (second heat)

Inherent Viscosity, acetone, 25°C = 0.753 dL/g Solutions of VF₂:HFP were spin coated at spin speeds of 3000 and 6000 rpm (revolutions per minute) onto CaF₂ substrates to produce polymer films of 69,800 angstroms and 1641 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 79:21 VF₂:HFP versus wavelength lambda (λ) in units of nanometers is shown in Figure 9 for Sample la. The 157 nm absorbance/micron determined from the thicker polymer film is 0.015/micron. The 193 nm absorbance/micron determined is 0.005/micron. The 248 nm absorbance/micron determined is 0.003/micron. EXAMPLE 6 - VF₂:HFP, VF₂:PDD, HFIB/TrFE pellicles

Using polymers such as VF₂:HFP, VF₂:PDD, and HFIB:TrFE, which have 157 nm absorbance/micron on the order of 0.01/micron, we can design tuned etalon pellicles with high transmissions and substantial film thicknesses. Figure 14 shows the transmission of a tuned etalon pellicle film of a polymer with an absorbance per micron of 0.01 and an index of refraction of 1.50 at a lithographic wavelength of 157 nm as a function of the pellicle film thickness. Note that for pellicle film thicknesses up to 8371 angstroms, the maximum pellicle transmission is above the target specification of 98%.

EXAMPLE 7 - TFE/TrP A. Polymerization of tetrafluoroethylene (TFE) with 3,3,3- trifluoropropene (TrP . A 240 ml autoclave was loaded with 20 ml of CF₂C1CC1₂F, chilled to < -20°C, and 10 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃ added. The autoclave was further chilled, evacuated, and 10 g of TrP and 40 g of TFE added. After shaking overnight at room temperature, the autoclave was vented, and the fluid reaction mixture evaporated down, finishing with 48 hours at room temperature under pump vacuum. This gave 9.96 g of sticky, colorless gum.

Elemental analysis, found: 29.86C; 1.74 H Calculated (C₃H₃F₃)₅(C₂H₄)₆: 30.02C; 1.40 H DSC, 10°C/min, nitrogen: Tg = 8.8°C (second heat)

Solution preparation and results:

A solution was prepared by rolling 4 g of poly (TFE/TrP) with 12 g of hexafluorobenzene and filtering through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™). The filtrate was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 5:6 TFP:TFE were spin coated onto CaF₂ substrates to produce polymer films of 41413 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 5:6 TFP:TFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 15 for Sample 17. The 157 nm absorbance/micron determined is 0.149/micron. The 193 nm absorbance/micron determined is 0.008/micron. The 248 nm absorbance/micron determined is -0.00085/micron.

B. Use as a Glue for Quartz and Aluminum. An aluminum coupon measuring 1" wide by 3" long by 122 mil thick and a quartz slide measuring 1 " wide by 3" long by 65 mil thick were both rinsed with CF₂C1CC1₂F and air dried.

Rolling 4 g of the poly(TFE/TrP) prepared above with 12 g of hexafluorobenzene at room temperature gave a clear, colorless solution that was passed though a 0.45μ glass fiber filter. Three drops of this solution were placed on one end of the aluminum coupon and a quartz slide pressed down on top so that the last inch of the quartz slide overlapped with the last inch of the aluminum coupon. This caused any excess fluid to squeeze out and the remaining polymer solution to wet and spread out evenly over the area of overlap. Two C-clamps (stock #72020, ACCO USA Inc., 770-S ACCO Plaza, Wheeling, IL 60090) were used to hold the aluminum coupon and quartz slide in place while the hexafluorobenzene solvent evaporated for three days at room temperature. One such assembly with C-clamps still attached was heated for 16 hours in a 50°C vacuum oven. The assembly was taken from the oven, cooled, the C-clamps removed, and the force required to pull the quartz slide away from the aluminum coupon measured in an Instron using a 3" jaw separation and a crosshead speed of 1 "/minute. Twelve pounds of force were required. A second such assembly with C-clamps still attached was heated for 20 hours in a 50°C vacuum oven and then for 24 hours in a 75°C vacuum oven. Now the Instron required 127 pounds of force to pull the aluminum coupon and the quartz slide apart.

C. Use as a Glue for a Pellicle Polymer . Pellicle polymer poly(HFIB/VF) was prepared as a film by dissolving 10 g of poly(HFIB/VF) (Example 2A below) in 40 g of 2-heptanone, adding 1 g of decolorizing carbon + 1 g of chromatographic alumina + 1 g of chromatographic silica gel, filtering through an 0.45 μ PTFE syringe filter, casting on Teflon™ FEP sheet with a 5 mil casting knife, and air drying. The poly(HFIB/VF) film prepared in this fashion could be lifted off the Teflon™ FEP sheet as clear, colorless film roughly 0.5 to 1 mil thick. Glue solution was prepared next. Solution was made by dissolving 0.1 g of the poly(TFE/TrP) prepared above in 1 g of hexafluorobenzene.

This glue solution was used to make several -1/2" spots on an aluminum coupon measuring 1 " wide by 3" long by 122 mil thick. These glue spots were air dried for 39 minutes and then the coupon was placed in a 60°C air oven for 8 minutes. As soon as the aluminum coupon was withdrawn hot from the oven, poly(HFIB/VF) film samples were pressed down on top of the poly(TFE/TrP) deposits with light finger pressure. The poly(HFIB/VF) film visibly wetted and adhered to the poly (TFE/TrP) spots. While the aluminum coupon was still hot, the poly(HFIB/VF) film could be pulled away with difficulty and stretching. Once, however, the aluminum coupon returned to room temperature, the poly(HFIB/VF) film tore rather than peeled free from the aluminum. The adhesive bond between the glue polymer [poly(TFE/TrP)] and the pellicle polymer [poly(HFIB/VF)] is quite strong, exceeding the strength of the pellicle polymer.

EXAMPLE 8 - HFIB/VF A. Polymerization of Hexafluoroisobutylene (HFIB-) with Vinyl Fluoride (VF) . A 400 ml autoclave was loaded with 200 ml water and 0.05 g of Vazo™ 56 WSP initiator. The autoclave was cooled and evacuated and then 80 g of hexafluoroisobutylene and 25 g of vinyl fluoride added. After shaking for -48 hours at 50°C, the autoclave was vented, and an opalescent blue emulsion recovered. The emulsion was broken by vigorous agitation in a Waring blender, filtered, and washed four times in the Waring blender with 100 ml of methyl alcohol. Drying for 6 days under pump vacuum gave 30 g of white powder and lumps.

Elemental analysis, found: 34.80% C; 2.57% H Calculated (C₄F₆H₂)₄₇(C₂H₃F)₅₃: 34.79C; 2.51% H DSC, 10°C/min, nitrogen: Tg = 70°C (second heat) Tm not detected

Inherent Viscosity, THF, 25°C: 0.379 dL/g GPC in THF: Mw = 192,000

Mn = 92,000 A sample of poly(HFIB/VF) was prepared a second time by the same emulsion polymerization method and its thermal transitions looked at in detail by modulated DSC. On the second heat, glass transitions were detected at both 69.5°C and 195°C. No melting transitions were observed on either first or second heats.

Solution preparation and results: A solution was made by rolling 10 g of the polymer with 40 g of 2-heptanone and filtering through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™). The filtrate was used to spin coat thick films on optical substrate for absorption measurements. Solutions of HFIB: VF were spin coated onto CaF₂ substrates to produce polymer films of 9239 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for HFIB:VF versus wavelength lambda (λ) in units of nanometers is shown in Figure 15 for

Sample 18. The 157 nm absorbance/micron determined is 0.005/micron. The 193 nm absorbance/micron determined is -0.00082/micron. The 248 nm absorbance/micron determined is -0.002/micron.

B. Use as a Glue for Quartz and Aluminum. An aluminum coupon measuring 1" wide by 3" long by 122 mil thick and a quartz slide measuring 1" wide by 3" long by 65 mil thick were both rinsed with CF₂C1CC1₂F and air dried.

Rolling 2 g of the poly(HFIB/VF) prepared above with 18 g of acetone at room temperature gave a clear, colorless solution that was passed though a 0.45 μ glass fiber filter. Three drops of this solution were placed on one end of the aluminum coupon and a quartz slide pressed down on top so that the last inch of the quartz slide overlapped with the last inch of the aluminum coupon. This caused any excess fluid to squeeze out and the remaining polymer solution to wet and spread out evenly over the area of overlap. Two C-clamps (stock #72020, ACCO USA Inc., 770-S ACCO Plaza, Wheeling, IL 60090) were used to hold the aluminum coupon and quartz slide in place while the acetone solvent evaporated for three days at room temperature. One such assembly with C-clamps still attached was heated for 16 hours in a 50°C vacuum oven. The assembly was taken from the oven, cooled, the C-clamps removed, and the force required to pull the quartz slide away from the aluminum coupon measured in an Instron using a 3" jaw separation and a crosshead speed of 1 "/minute. One pound of force was required. A second such assembly with C-clamps still attached was heated for 16 hours in a 50°C vacuum oven and then for 24 hours in a 75°C vacuum oven. Now the Instron required 0.6 pounds of force to pull the aluminum coupon and the quartz slide apart.

EXAMPLE 9 - VF2/PMVE

A. Polymerization of Vinylidene Fluoride CVF2) with Perfluoromethylvinyl Ether fPMVE . A -210 ml autoclave was loaded with 50 ml of CF₂C1CC1₂F and 10 ml of -0.2 DP in CF₃CFHCFHCF₂CF₃ solvent. The autoclave was cooled evacuated and 26 g VF2 and 33 g PMVE added. After shaking overnight at ambient temperatures (18-26°C), the autoclave was vented, and viscous solution recovered. Excess solvent was evaporated under nitrogen and the polymer dried 72 hours under vacuum at room temperature and then for 28 hours at 75°C in a vacuum oven, 33 g. Elemental analysis, found: 29.64% C 1.73% H

Calculated (C₂F₂H₂)₅(C₃F₆0)₂: 29.47% C 1.55 % H DSC, 10°C/min, nitrogen: Tg = -32°C (second heat) Inherent Viscosity, CF₃CFHCFHCF₂CF₃,25°C: 0.722 dL/g Solution preparation and results: A solution was made by rolling 4 g of this polymer with with 16 g of hexafluorobenzene. This solution was passed through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™) and the filtrate used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 5:2 VF2:PMVE were spin coated onto CaF₂ substrates to produce polymer films of 72,750 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for 5:2 VF2:PMVE versus wavelength lambda (λ) in units of nanometers is shown in Figure 15 for Sample 19. The 157 nm absorbance/micron determined is 0.016/micron. The 193 nm absorbance/micron determined is 0.006/micron. The 248 nm absorbance/micron determined is 0.004/micron.

B. Use as a Glue for Quartz and Aluminum. An aluminum coupon measuring 1" wide by 3" long by 122 mil thick and a quartz slide measuring 1" wide by 3" long by 65 mil thick were both rinsed with CF₂C1CC1₂F and air dried. About 0.2 g of the poly(VF2/PMVE) was chopped into 3 to 4 small pieces and placed on one end of the aluminum coupon. A quartz slide was pressed down on top so that the last inch of the quartz slide overlapped with the last inch of the aluminum coupon. Two C-clamps (stock #72020, ACCO USA Inc., 770-S ACCO Plaza, Wheeling, IL 60090) were used to hold the aluminum coupon and quartz slide in place and the assembly with C-clamps still attached placed in a 76-80°C vacuum oven for 22 hours causing the polymer to spread out as a clear, colorless layer, excess polymer extruding out the edges. The assembly was taken from the oven, cooled, the C-clamps removed, and the force required to pull the quartz slide away from the aluminum coupon measured in an Instron using a 3" jaw separation and a crosshead speed of 1 "/minute. A force of 105.3 lbs was required. Separation occurred in the polymer layer, residual polymer being retained on both the aluminum and the quartz. A second similarly prepared sample required 101.0 pounds of force to pull apart for an average force of 103 lbs.

EXAMPLE 10 - VF2/PPVE A 210 ml autoclave was loaded with 50 ml of CF₂C1CC1₂F, chilled to <-20°C, and 10 ml of -0.2 M DP in CF₃CFHCFHCF₂CF₃ added. The autoclave was evacuated and loaded with 33 g perfluoro(methyl vinyl ether) (PMVE) and 13 g vinylidene fluoride (VF2). Shaking the autoclave overnight at room temperature gave solution that was evaporated down to soft polymer and then further dried for 29 hours under pump vacuum and for 20 hours in a 75 °C vacuum oven. This gave 31 g of soft polymer suitable for use as a glue. Calc. for (C₂H₂F₂)₁₃(C₃F₆O)₁₀: 26.98% C 1.05% H

Found: 27.21% C 0.88% H

DSC, 10°C/min, nitrogen: Tg = -29°C (second heat)

Inherent Viscosity, 2-heptanone: 0.166 dL/g

Solution preparation and results: A solution was made by rolling 4 g of this polymer with with 16 g of hexafluorobenzene. This solution was passed through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™) and the filtrate used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 13:10 VF2:PMVE were spin coated onto CaF₂ substrates to produce polymer films of 25,970 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for 13:10 VF2:PMVE versus wavelength lambda (λ) in units of nanometers is shown in Figure 16 for Sample 20. The 157 nm absorbance/micron determined is 0.034/micron. The 193 nm absorbance/micron determined is 0.015/micron. The 248 nm absorbance/micron determined is 0.018/micron.

EXAMPLE 11 - VF2/PPVE A. Polymerization of Vinylidene Fluoride (VF2 with PerfluoroCpropyl Vinyl Ether (PPVEV A 210 ml autoclave was loaded with 50 ml of CF₂C1CC1₂F, chilled to <-20°C, and 10 ml of -0.2 M DP in CF₃CFHCFHCF₂CF₃ added. The autoclave was cooled evacuated and 13 g VF2 and 53 g PPVE added. After shaking overnight at ambient temperatures (22-26°C), the autoclave was vented, and solution recovered. Excess solvent was evaporated under nitrogen and the polymer dried 72 hours under vacuum at room temperature and then for 28 hours at 75°C, 45 g.

Elemental analysis, found: 26.17% C; 0.88% H Calculated (C₂F₂H₂)₇(C₅F₁₀O)₅: 26.34% C; 0.79 % H DSC, 10°C/min, nitrogen: Tg = -32°C (second heat)

Inherent Viscosity, hexafluorobenzene, 25°C: 0.169 dL/g

Solutions of 7:5 VF2:PFPVE were spin coated onto CaF₂ substrates to produce polymer films of 29,874 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 7:5 VF2:PFPVE versus wavelength lambda (λ) in units of nanometers is shown in Figure 15 for Sample 21 . The 157 nm absorbance/micron determined is 0.028/micron. The 193 nm absorbance/micron determined is -0.003/micron. The 248 nm absorbance/micron determined is -0.00074/micron. B. Use as a Glue for a Pellicle Polymer. Poly(HFIB/VF) was prepared as a polymer film by dissolving 10 g of poly(HFIB/VF) (Example 8 above) in 40 g of 2-heptanone, adding 1 g of decolorizing carbon + 1 g of chromatographic alumina + 1 g of silica gel, filtering through an 0.45 μ PTFE syringe filter, casting on Teflon™ FEP sheet with a 5 mil casting knife, and air drying. The poly(HFIB/VF) film prepared in this fashion could be lifted off the Teflon™ FEP sheet as clear, colorless film roughly 0.5 to 1 mil thick.

Glue solution was prepared next. Solution was made by dissolving 0.1 g of the poly(VF2/PPVE) prepared above in 1 g of hexafluorobenzene. This glue solution was used to make a -1/2" spot on an aluminum coupon measuring 1" wide by 3" long by 122 mil thick. This glue spot was air dried for 39 minutes and then the coupon was placed in a 62°C oven under nitrogen for 13 minutes. As soon as the aluminum coupon was withdrawn hot from the oven, a poly(HFIB/VF) film sample was pressed down on top of the poly(VF2/PPVE) deposit with light finger pressure. Once, the aluminum coupon returned to room temperature, the poly(HFIB/VF) film tore when either pulled as in a lap shear or peeled back. The adhesive bond between the glue polymer [poly(VF2/PPVE)] and the pellicle polymer [poly(HFIB/VF)] was quite strong, exceeding the strength of the pellicle polymer. EXAMPLE 12 - VF2/HFP

A. PolvCvinylidene fluoride/hexafluoropropylene TVF2/HFP Sample. A vinylidene fluoride/hexafluoropropylene sample made by the method of US Patent 4,985,520 of 1/15/91 was characterized. Composition by Fluorine NMR in Deuteroacetone: 21 mole % VF2, 79 mole % HFP

DSC, 10°C/min, nitrogen: Tg = -22.5°C (second heat) Inherent Viscosity, acetone, 25°C: 0.753 dL/g

Solution preparation and results: A solution was made by rolling 3 g of poly(VF2/HFP) with 17 g of 2-heptanone and filtering through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™). The filtrate used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 79:21 VF2:HFP were spin coated onto CaF₂ substrates to produce polymer films of 13,000 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for 79:21 VF2:HFP versus wavelength lambda (λ) in units of nanometers is shown in Figure 15 for sample 22. The 157 nm absorbance/micron determined is 0.014/micron. The 193 nm absorbance/micron determined is -0.002/micron. The 248 nm absorbance/micron determined is -0.00056/micron.

About 0.2 g of the poly(VF2/HFP) was chopped into 3 to 4 small pieces and placed on one end of the aluminum coupon. A quartz slide was pressed down on top so that the last inch of the quartz slide overlapped with the last inch of the aluminum coupon. Two C-clamps (stock #72020, ACCO USA Inc., 770-S ACCO Plaza, Wheeling, IL 60090) were used to hold the aluminum coupon and quartz slide in place and the assembly with C-clamps still attached placed in a 76-80°C vacuum oven for 22 hours causing the polymer to spread out as a clear, colorless layer. The assembly was taken from the oven, cooled, the C-clamps removed, and the force required to pull the quartz slide away from the aluminum coupon measured in an Instron using a 3" jaw separation and a crosshead speed of 1 "/minute. A force of 133.3 lbs was required. A second similarly prepared sample required 121.6 pounds of force to pull apart for an average force of

127 lbs. In second of the two experiments the separation occurred in the polymer layer, residual polymer being retained on both the aluminum and the quartz. In the first of the two experiments the separation occurred between the polymer and the glass. In either event, residual polymer could be pulled cleanly off the glass and the aluminum as a single piece.

C. Use as a Glue for a Pellicle Polymer. Poly(HFIB/VF) was prepared as a polymer film by dissolving 10 g of poly(HFIB/VF) (Example 2 A above) in 40 g of 2-heptanone, adding 1 g of decolorizing carbon + 1 g of chromatographic alumina + 1 g of silica gel, filtering through an 0.45 μ PTFE syringe filter, casting on Teflon™ FEP sheet with a 5 mil casting knife, and air drying. The poly(HFIB/VF) film prepared in this fashion could be lifted off the Teflon™ FEP sheet as clear, colorless film roughly 0.5 to 1 mil thick. Glue solution was prepared next. Solution was made by dissolving 0.1 g of the poly(VF2/HFP) prepared above in 1 g of hexafluorobenzene.

This glue solution was used to make a -1/2" spot on an aluminum coupon measuring 1 " wide by 3" long by 122 mil thick. This glue spot was air dried for 39 minutes and then the coupon was placed in a 62°C nitrogen blanketed oven for 13 minutes. As soon as the aluminum coupon was withdrawn hot from the oven, a poly(HFIB/VF) film sample was pressed down on top of the poly(VF2/HFP) deposit with light finger pressure. Once, the aluminum coupon returned to room temperature, the poly(HFIB/VF) film tore when pulled as in a lap shear. When the poly(HFIB/VF) film was peeled back on itself, adhesive failure occurred between the poly(VF2/HFP) glue and the aluminum, the poly(VF2/HFP) adhesive layer coming away clean from the aluminum and showing no debonding from the poly(HFIB/VF). This shows excellent adhesion between the pellicle polymer [poly(HFIB/VF)] and the glue polymer [poly(VF2/HFP)] .

EXAMPLE 13 - HFP/PMVE/VF2 A 210 ml Hastelloy C autoclave was chilled to <-20°C and loaded with

20 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃ added. The autoclave was evacuated and loaded with 33 g perfluoro(methyl vinyl ether) (PMVE), 26 g vinylidene fluoride (VF2), 26 g of hexafluoropropylene (HFP), and 1 g of anhydrous hydrogen chloride (chain transfer agent). Shaking the autoclave overnight at room temperature gave colorless oil that was devolatilized for

24 hours in air, 19 hours under pump vacuum, and then for five days in a 75°C vacuum oven, giving 47 g of clear, tacky resin suitable for use as a glue. Composition by Fluorine NMR: 63.1 mole % VF2

27 mole % PMVE 10 mole % HFP

Inherent viscosity, acetone, 25°C: 0.120 dL/g

Solution preparation and results: A solution was made by rolling 4 g of this polymer with with 12 g of 2-heptanone to give a hazy solution. This solution was passed through a 0.45 μ glass microfiber and Teflon™ syringe filters (Whatman, Autovial™) and the filtrate used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 10:27:63 HFP:PMVE:VF2 were spin coated onto CaF₂ substrates to produce polymer films of 316,500 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 10:27:63 HFP:PMVE:VF2 versus wavelength lambda (λ) in units of nanometers is shown in Figure 16 for Sample 23. The 157 nm absorbance/micron determined is 0.008/micron. The 193 nm absorbance/micron determined is -0.00048/micron. The 248 nm absorbance/micron determined is -0.00045/micron.

EXAMPLE 14 - HFIB/VF An 85 ml autoclave was loaded with two 3/8" diameter stainless steel balls, sealed, evacuated, chilled to <-20°C, and 10 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃ sucked in. The autoclave was further chilled and 70 g hexafluoroisobutylene (HFIB) added at — 50°C and 10 psig liquid vinyl fluoride (VF) added at — 35°C. The now liquid full autoclave was shaken and allowed to warm to room temperature. Pressure in the clave maxed out at 7780 psi at 25°C, finishing at 103 psig at 29°C about 10 hours into the run. Drying the solid polymeric product for 72 hours under pump vacuum gave 57 g of white solid with a glass transition temperature of 115°C. Calc. for (HFIB)₅₉(VF)₄₁: 33.04% C 2.11% H

: 33.02% C 2.31% H

DSC, 10°C/min, nitrogen : Tg = 115°C (second heat) Tm not detected (first or second heats)

Inherent Viscosity, THF, 25°C: 0.183 dL/g

Solution preparation and results: A solution was made by rolling 12 g of this polymer with with 48 g of 2-heptanone. This solution was passed through a 0.45 μ glass microfiber and Teflon™ syringe filters (Whatman, Autovial™) and the filtrate used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 10:7 HFIB:VF were spin coated onto CaF substrates to produce polymer films of 7500 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 10:7 HFIB:VF versus wavelength lambda (λ) in units of nanometers is shown in Figure 16 for Sample 24. The 157 nm absorbance/micron determined is -0.013/micron. The 193 nm absorbance/micron determined is -0.016/micron. The 248 nm absorbance/micron determined is -0.011/micron.

EXAMPLE 15 - PDD/PMVE A 75 ml autoclave was loaded with 25 ml of CF₂C1CC1₂F, chilled to <-20°C, and 10 ml of -0.2 M DP in CF₃CFHCFHCF₂CF₃ and 11.6 ml of perfluorodimethyldioxole (PDD) added. The autoclave was evacuated and loaded with 4.5 g perfluoro(methyl vinyl ether) (PMVE). Shaking the autoclave overnight at room temperature gave a gelatinous product that was evaporated down and then further dried for 29 hours under pump vacuum and for 20 hours in a 75°C vacuum oven. This gave 18 g of product.

Calc. for (PDD)₆(PMVE)₅: 23.56% C

Found: 23.87% C

DSC, 10°C/min, nitrogen: Tg = 133°C

Inherent Viscosity, Fluorinert™ FC-40: 0.161 dL/g Solution preparation and results: A hazy solution was made by rolling 4 g of this polymer with with 16 g of Fluorinert™ FC-40. This solution was passed through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™) and then, after the addition of 0.6 g of chromatographic alumina and 0.6 g of chromatographic silica gel, through a 0.45 μ Teflon™ syringe filter (Whatman, Autovial™). The filtrate was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 6:5 PDD:PFMVE were spin coated onto CaF₂ substrates to produce polymer films of 12,450 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 6:5 PDD:PFMVE versus wavelength lambda (λ) in units of nanometers is shown in Figure 16 for Sample 25. The 157 nm absorbance/micron determined is 0.209/micron. The 193 nm absorbance/micron determined is 0.006/micron. The 248 nm absorbance/micron determined is 0.013/micron. EXAMPLE 16 - VF/CIDE

A 75 ml autoclave was loaded with 25 ml of CF₂C1CC1₂F, chilled to <- 20°C, and 10 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃ added. The autoclave was evacuated and loaded with 6.4 g of vinyl fluoride (VF) and 9.8 g 1-chloro- 2,2-difluoroethylene (C1DFE). Shaking the autoclave overnight at room temperature gave damp paste that was dried for 4 days under pump vacuum and for 28 hours in a 75°C vacuum oven. This gave 8 g of product. Calc. for (C₂H₃F)₂₀(C₂HF₂Cl)_π: 37.16% C 3.57% H Found: 37.39% C 3.30% H

DSC, 10°C/min, nitrogen: Tg = 97°C (first heat)

Tg not detected second heat Inherent Viscosity, acetone: 0.220 dL/g

Solution preparation and results: A solution was made by rolling 4 g of this polymer with with 16 g of heptanone. This solution was passed through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™) and the filtrate used to spin coat thick films on optical substrate for absorption measurements. Solutions of 20: 11 VF:C1DFE were spin coated onto CaF₂ substrates to produce polymer films of 9461 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for 20:11 VF:C1DFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 17 for Sample 26. The 157 nm absorbance/micron determined is 0.226/micron. The 193 nm absorbance/micron determined is 0.036/micron. The 248 nm absorbance/micron determined is 0.003/micron.

EXAMPLE 17 - PDD/VF2 A 75 ml autoclave was chilled to <-20°C and loaded with 10 ml of CF₃CFHCFHCF₂CF₃, 5 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃, and 20 g of perfluorodimethyldioxole (PDD). The autoclave was evacuated and loaded with 13 g vinylidene fluoride (VF2). Shaking the autoclave overnight at room temperature gave viscous oil and that was devolatilized in air, 32 hours under pump vacuum, and then for 23 hours in a 75°C vacuum oven, giving 16 g of white resin.

Calc. for (PDD)!(VF2)₂: 29.05% C 1.08% H

Found: 29.35% C 1.08% H

DSC, 10°C/min, nitrogen: 52°C? (weak)

Inherent Viscosity, hexafluorobenzene, 25°C: 0.278 Solution preparation and results: A solution was made by rolling 2 g of poly(PDD/VF2) with with 8 g of 2-hexafluorobenzene. This solution was passed through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™) and the filtrate used to spin coat thick films on optical substrate for absorption measurements. Solutions of 1 :2 PDD:VF2 were spin coated onto CaF₂ substrates to produce polymer films of 82,000 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 1 :2 PDD:VF2 versus wavelength lambda (λ) in units of nanometers is shown in Figure 17 for Sample 27. The 157 nm absorbance/micron determined is 0.009/micron. The 193 nm absorbance/micron determined is 0.003/micron. The 248 nm absorbance/micron determined is -0.00030/micron.

EXAMPLE 18A - PDD/VF2 A 75 ml autoclave was chilled to <-20°C and loaded with 20 ml of CF₃CFHCFHCF₂CF₃, 5 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃, and 20 g of perfluorodimethyldioxole (PDD). The autoclave was evacuated and loaded with 6 g vinylidene fluoride (VF2). Shaking the autoclave overnight at room temperature gave viscous oil and that was devolatilized in air, 32 hours under pump vacuum, and then for 23 hours in a 75°C vacuum oven, giving 16 g of white resin.

Calc. for (C₅F₈O2)2(C₂H2F2)i: 26.11% C 0.37% H Found: 26.03% C 0.53% H

DSC, 10°C/min, nitrogen: 96°C? (weak)

Inherent Viscosity, hexafluorobenzene, 25°C: 0.671

Solution preparation and results: A hazy solution was made by rolling 2 g of poly(PDD/VF2) with with 18 g of lH-perfluorohexane. This solution was passed through an 0.45 μ glass microfiber syringe filter (Whatman, Autovial™) with the aid of a pad of chromatographic silica gel to help in the filtration, additional 2-H-perfluorohexane added to approximately replace that lost to evaporation,and then filtered again through an 0.45 μ Teflon™ syringe filter (Whatman, Autovial™). The now clear filtrate was used to spin coat thick films on optical substrate for absorption measurements.

EXAMPLE 18B - PDD/VF2 A Teflon™ coated 400 ml stainless steel autoclave chilled to <-20°C was loaded with 90 ml of CF₃CFHCFHCF₂CF₃, 30 ml of -0.1 M DP in CF₃CFHCFHCF CF3, and 120 g of perfluorodimethyldioxole. The autoclave was pressured to 100 psi with nitrogen and evacuated ten times. Finally 58 g of vinylidene fluoride (VF2) were added and the autoclave shaken overnight at room temperature. The resulting thick gel was dried under nitrogen, then under pump vacuum, and finally for 4 days in a 75 °C vacuum oven, giving 122 g of white resin. Calc. for (C₅F₈O₂)₅(C₂H₂F₂)₈: 28.42% C 0.93% H

Found: 28.22% C 1.16% H

DSC, 10°C/min, nitrogen 59°C

Inherent Viscosity, hexafluorobenzene, 25°C: 0.576 Solution preparation: A solution was made by rolling 1.70 g of the poly(PDD/VF2) with 15.3 g of hexafluorobenzene. Filtration through a 0.45 μ glass fiber microfiber syringe filter (Whatman Autovial™) gave clear, colorless solution. The filtrate was used to spin coat films on optical substrate for absorption measurements.

Solutions of 5:8 PDD:VF2 were spin coated onto CaF₂ substrates to produce polymer films of 38,298 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for 5:8 PDD:VF2 versus wavelength lambda (λ) in units of nanometers is shown in Figure 19 for

Sample 29. The 157 nm absorbance/micron determined is 0.018/micron. The 193 nm absorbance/micron determined is 0.010/micron. The 248 nm absorbance/micron determined is 0.000057/micron.

EXAMPLE 19A - HFIB/VF2/VF A 210 ml Hastelloy C autoclave was loaded with 50 ml of CF₂C1CC1₂F, chilled to <-20°C, and 15 ml of -0.2 M DP in CF₃CFHCFHCF₂CF₃ added. The autoclave was evacuated and loaded with 32 g hexafluoroisobutylene (HFIB), 3 g vinylidene fluoride (VF2), and 7 g vinyl fluoride (VF). Shaking the autoclave overnight at room temperature gave a thick paste of white solids that was dried under nitrogen, for 70 hours under pump vacuum, and for 23 hours in a 75°C vacuum oven. This gave 33 g of product having an analysis matching a range of possible stoichiometries.

Calc. for (C4H2F₆)3(C2H3F)₁(C₂H2F2)i: 31.91% C 1.84% H

Calc. for (C4H₂F₆)5(C2H3F)₁(C₂H₂F2)3: 31.78% C 1.81% H Found: 31.78% C 1.88% H

DSC, 10°C/min, nitrogen: Tg = 48°C

Inherent Viscosity, acetone, 25°C: 0.030 dL/g

Solution preparation and results: A hazy solution was made by rolling 8 g of this polymer with with 20 g of heptanone. This solution was passed through a 0.45 μ glass fiber syringe filter (Whatman, Autovial™), a 0.45 μ Teflon™ syringe filter (Whatman, Autovial™), and a 0.45 μ Teflon™ syringe filter (Whatman, Autovial™) with the aid of a 1 g pad of chromatographic alumina to aid in the filtration. The now clear filtrate was used to spin coat thick films on optical substrate for absorption measurements. EXAMPLE 19B - HFIB/VF/VF2

A 400 ml autoclave was loaded with 200 ml deionized water, 0.05 g Vazo™ 56 WSP initiator, and pressured to 100 psi with nitrogen and evacuated ten times. The autoclave was chilled to <-20°C, evacuated, and further loaded with 82 g of hexafluoroisobutylene (HFIB), 13 g of vinylidene fluoride (VF2), and 14 g of vinyl fluoride (VF). The contents of the autoclave were shaken for -64 hours at 50°C. The resulting mix of solids plus emulsion was frozen and thawed. Vacuum filtration and washing twice in the filter with 200 ml of water gave white fluffy solid with some darker yellow chunks that were removed. The remaining white fluff was washed twice more in the filter with 150 ml of methyl alcohol. Sucking the white fluff dry on the filter and then further drying for 5 days under pump vacuum gave 16 g of polymer. Composition by fluorine NMR: 41 mole % HFIB, 37 mole % VF, 22 mole % VF2

DSC, 10°C/min, nitrogen 71°C

Inherent Viscosity, tetrahydrofuran, 25°C: 0.523

Solution preparation: A solution was made by rolling 3 g of poly(HFIB/VF/VF2) with 17 g of 2-heptanone. . Filtration through a 0.45 μ glass fiber microfiber syringe filter and then through a 0.45 μ PTFE syring filter

(Whatman Autovial™) gave a clear but faintly yellow solution. This filtrate was used to spin coat films on optical substrate for absorption measurements. Solutions of 41:37:22 HFIB:VF:VF2 were spin coated onto CaF₂ substrates to produce polymer films of 5289 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 41:37:22 HFIB:VF:VF2 versus wavelength lambda (λ) in units of nanometers is shown in Figure 19 for Sample 31. The 157 nm absorbance/micron determined is 0.016/micron. The 193 nm absorbance/micron determined is -0.010/micron. The 248 nm absorbance/micron determined is -0.002/micron.

EXAMPLE 20 - PDD/TrFE A 75 ml autoclave was loaded with 40 ml water and 0.1 g of Vazo™ 56 WSP initiator. The autoclave was chilled to <-20°C and 10 g of perfluorodimethyldioxole (PDD) added. The autoclave was evacuated and 5 g of trifluoroethylene (TrFE) added. Shaking for 10 hours at 70°C gave a lump of polymer that was separated and dried for 72 hours in a 75°C vacuum oven, 9.9 g. Elemental analysis, found: 25.51 % C 0.55% H

Calculated (C₅F₈O2)₁(C₂F3H)₁: 25.79C 0.31% H

DSC, 10°C/min, nitrogen: Tg = 150°C? (weak, second heat) Inherent Viscosity, Hexafluorobenzene, 25°C: 1.3 dL/g

Solution preparation and results: A solution was made by rolling 1 g of the polymer with 33 g of Fluorinert™ Fc-75 and filtering through a 0.45 μ glass microfiber syringe filter (Whatman, Autovial™). The filtrate was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 1:1 PDD:TrFE were spin coated onto CaF₂ substrates to produce polymer films of 7688 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for 1:1 PDD:TrFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 16 for Sample 9. The 157 nm absorbance/micron determined is 0.03/micron. The 193 nm absorbance/micron determined is -0.004/micron. The 248 nm absorbance/micron determined is -0.001/micron.

EXAMPLE 21 - CTFE/VF A 75 ml autoclave was loaded with 25 ml of CF₂C1CC1₂F, chilled to <-20°C, and 5 ml of -0.1 M DP in CF₃CFHCFHCF₂CF₃ added. The autoclave was evacuated and loaded with 17 g of chlorotrifluoroethylene (CTFE) and 7 g vinyl fluoride (VF). Shaking the autoclave overnight at room temperature gave a swollen gel that was dried in air, for 96 hours under pump vacuum, and for 25 hours in a 75 °C vacuum oven. This gave 18 g of polymer. Calc. for (C2F₃Cl)₁(C2H₃F)₁: 29.56% C 1.86% H 21.82% Cl

Found: 29.74% C 1.83% H 21.80% Cl DSC, 10°C/min, nitrogen: Tg = 86°C (first heat)

Inherent Viscosity, acetone, 25°C: 1.08 dL/g

Solution preparation and results: A solution was made by rolling 2 g of this polymer with with 15 g of heptanone. This solution was passed through a 0.45 μ glass fiber syringe filter (Whatman, Autovial™) and the filtrate used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 1:1 CTFE:VF were spin coated onto CaF₂ substrates to produce polymer films of 1850 angstroms thickness and 17,644 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 1 : 1 CTFE: VF versus wavelength lambda (λ) in units of nanometers is shown in Figure 18 for Sample 6b. The 157 nm absorbance/micron determined from the thicker film is 0.388/micron. The 193 nm absorbance/micron determined from the thicker film is 0.016/micron. The 248 nm absorbance/micron determined from the thicker film is 0.006/micron.

SAMPLE 10 - VF2/TrFE A 75 ml autoclave was loaded with 25 ml of CF₂C1CC1 F, chilled to <-20°C, and 5 ml of -0.1 M DP in CF₃CFHCFHCF₂CF₃ added. The autoclave was evacuated and loaded with 9.6 g of vinylidene fluoride (VF2) and 12 g trifluoroethylene (TrFE). Shaking the autoclave overnight at room temperature gave a damp white solid that was dried under pump vacuum and then 24 hours in a 75 °C vacuum oven. This gave 14 g of polymer. Calc. for (C₂H₃F₂)₅(C₂F₃H)₂: 34.73% C 2.50% H

Found: 34.78% C 2.48% H

DSC, 10°C/min, nitrogen: first heat, Tg = 55°C second heat, Tg = 98 and 147°C Inherent Viscosity, acetone, 25°C: 1.087 dL/g Solution preparation and results: A solution was made by rolling 3.11 g of this polymer with with 34.32 g of l-methoxy-2-propanol acetate. The resulting hazy solution with particulates passed with difficulty through an 0.45 μ glass fiber syringe filter (Whatman, Autovial™). Evaporation of 2.82 g of this solution gave 0.220 g of residue (7.8 wt % solids vs. 8.3% expected if no solids eliminated by filtration). Thick films were spin coated on optical substrate for absorption measurements.

Solutions of 5:2 VF2:TrFE were spin coated onto CaF₂ substrates to produce polymer films of4500 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 5 :2 VF2:TrFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 18 for Sample 10. The 157 nm absorbance/micron determined is 0.924/micron. The 193 nm absorbance/micron determined is 0.188/micron. The 248 nm absorbance/micron determined is 0.083/micron. EXAMPLE 23 - HFIB/VOH

A. Hexafluoroisobutylene (ΗFIBWinyl Acetate (VOAc) Copolymer . A 400 ml autoclave was loaded with 50 ml of CF₂C1CC1₂F, 27 ml of vinyl acetate, and 50 ml of -0.1M DP in CF₃CFHCFHCF₂CF₃. The autoclave was chilled, evacuated, and futher loaded with 49 g of HFIB. Shaking overnight at room temperature gave a viscous pale yellow solution. The polymer was precipitated by addition to 400 ml of methyl alcohol. Filtration, drying under pump vacuum, and drying for 19 hours in a 75°C vacuum oven gave 52 g of poly(HFIB/VOAc) having an inherent viscosity of 0.13 in acetone at 25°C.

B. Hexafluoroisobutylene (HFIB inyl Alcohol (VOH) Copolymer. Fifty grams of the poly(HFIB/VOAc) prepared above, 500 ml of methyl alcohol, and 25 ml of 25 wt % sodium methoxide in methyl alcoh ol were refluxed for 6 hours. Over this period additional methyl alcohol was added as -860 ml of methyl alcohol were distilled off. Addition of resulting polymer solution to water gave a gummy precipitate. The water was decanted off and the precipitate taken back into 200 ml of methyl alcohol and reprecipitated with 500 ml of water in a Waring blender. Vacuum filtration, drying under pump vacuum and then for 24 hours in a 75 °C vacuum oven gave 37 g of poly(HFIB/VOH) as off white granules.

Calc. for 34.63% C 2.91% H

Found: 34.34% C 2.73% H

DSC, 10°C/min, nitrogen: Tg = 90°C (second heat)

Solution preparation and results: A solution was made by rolling 3.28 g of this polymer with with 34.91 g of l-methoxy-2-propanol acetate. This solution was passed through a 0.45 μ glass fiber syringe filter (Whatman, Autovial™) giving a slightly hazy pale yellow solution. The solution was treated with 0.4 g decolorizing carbon and filtered again and then with 0.58 g of silica gel and filtered a third time. The solution was still pale yellow but no longer hazy. The filtrate used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 1 :1 HFIB:VA were spin coated onto CaF₂ substrates to produce polymer films of 1350 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 1 : 1 HFIB:VA versus wavelength lambda (λ) in units of nanometers is shown in Figure 17 for Sample 32. The 157 nm absorbance/micron determined is 0.350/micron. The 193 nm absorbance/micron determined is -0.047/micron. The 248 nm absorbance/micron determined is -0.107/micron. SAMPLE 12 - HFP/TrFE

A 75 ml autoclave was loaded with 25 ml of CF₂C1CC1₂F, chilled to <-20°C, and 5 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃ added. The autoclave was evacuated and loaded with 12 g hexafluoropropylene (HFP) and 12 g of trifluoroethylene (TrFE). Shaking the autoclave overnight at room temperature gave a damp white solid that was dried for 22 hours under pump vacuum and for 24 hours in a 75°C vacuum oven. This gave 12 g of product. Composition by Fluorine NMR: 2 mole % HFP, 98 mole % TrFE DSC, 10°C/min, nitrogen: Tm = 179°C, Tg not detected

Inherent Viscosity, acetone, 25°C: 0.912 dL/g Solution preparation: Rolling 2.5 g of poly(HFP/TrFE) with 50 ml of propylene glycol methyl ether acetate gave a hazy solution. Filtration though an 0.45 μ PTFE syringe filter (Whatman, Autovial™) gave clear filtrate. The filtrate was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 2:98 HFP:TrFE were spin coated onto CaF₂ substrates to produce polymer films of 1389 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for HFP: TrFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 8 for Sample 12. The 157 nm absorbance/micron determined is 1.37/micron. The 193 nm absorbance/micron determined is 0.143/micron. The 248 nm absorbance/micron determined is -0.02/micron.

SAMPLE 14 - HFP/TFE Hexafluoropropylene(HFP)/tetrafluoroethylene(TFE) copolymer was prepared by the method of US 5,478,905 of December 26, 1995. This polymer was 60 wt % HFP, 40 wt % TFE, and had an inherent viscosity of 0.407 dL/g (in Fluorinert™ FC-75 solvent at 25°C).

Solution preparation: Rolling 31.6 g of HFP/TFE copolymer with 986 g of PF-5080 (Performance Fluid manufactured by 3M, believed to be largely perfluorooctanes) gave solution. Vacuum filtration through an 0.45 μ filter gave clear solution. The filtrate was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 1 :1 HFP:TFE were spin coated onto CaF₂ substrates to produce polymer films of 1850 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for 1:1 HFP:TFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 8 for Sample 14. The 157 nm absorbance/micron determined is 3.9/micron. The 193 nm absorbance/micron determined is 0.086/micron. The 248 nm absorbance/micron determined is 0.073/micron.

SAMPLE 15 - VF2/CTFE A 75 ml autoclave chilled to <-20°C was loaded with 25 ml of

CF₂C1CC1₂F and 5 ml of -0.1 M DP in CF₃CFHCFHCF₂CF₃. The autoclave was evacuated and loaded with 9.6 g of vinylidene fluoride (VF2) and 17 g of chlorotrifluoroethylene (CTFE). Shaking the autoclave overnight at room temperature gave viscous white fluid that was evaporated down to stretchy solid and then dried for for 1 hour under pump vacuum and for 30 hours in a 75°C vacuum oven. This gave 19 g of product. Calc (C₂F2H2)5(C₂F₃C1)₄: 27.50% C 1.28% H 18.04% Cl

Found: 27.26% C 1.41% H 17.91% Cl

DSC, 10°C/min, nitrogen: Tg = 99°C

Inherent Viscosity, acetone, 25°C: 0.546 dL/g Solution preparation: Rolling 3 g poly(VF2/CTFE) with 20 g of

2-heptanone gave solution that was filtered though an 0.45 μ glass fiber syringe filter (Whatman, Autovial™). The filtrate was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 5:4 VF2:CTFE were spin coated onto CaF₂ substrates to produce polymer films. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for 5:4 VF2:CTFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 18 for Sample 15. The 157 nm absorbance/micron determined is 5.6/micron. The 193 nm absorbance/micron determined is 0.27/micron. The 248 nm absorbance/micron determined is 0.12/micron.

SAMPLE 16 - PMD/TFE A 75 ml autoclave chilled to <-20°C was loaded with 5 ml of -0.14 M DP in CF₃CFHCFHCF₂CF₃, 25 ml of CF₃CFHCFHCF₂CF₃, and 11.6 ml of perfluoro(2-methylene-4-methyl-l,3-dioxolane) (PMD). The tube was evacuated and further loaded with 5 g of tetrafluoroethylene (TFE). Shaking overnight at room temperature gave a milky, viscous oil. Evaporation under nitrogen, 76 hours under pump vacuum, and 24 hours in a 75°C vacuum oven gave 20 g of resin assumed to be a -2:1 copolymer of PMD:TFE (monomers mixed in ~2PMD:1 TFE ratio).

DSC, 10°C/min, nitrogen: Tg not detected

Inherent Viscosity, Fluorinert™ FC-75, 25°C: 0.142 dL/g

Solution preparation: Rolling 2.14 g of poly(PMD/TFE) with 26.2 g of Fluorinert™ FC-40 gave a hazy solution that was passed through an 0.45 μ glass fiber filter (Whatman, Autovial™), through an 0.45 μ glass fiber syringe filter

(Whatman, Autovial™) after mixing with 0.2 g chromatographic silica + 0.2 g of decolorizing carbon, and finally through an 0.45 μ PTFE syringe filter (Whatman, Autovial™). The filtrate was used to spin coat thick films on optical substrate for absorption measurements. Solutions of 1 : 1 PMD:TFE were spin coated onto CaF₂ substrates to produce polymer films of 2207 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron. The absorbance in units of inverse microns for 1:1 PMD:TFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 18 for Sample 16. The 157 nm absorbance/micron determined is 1.17/micron. The 193 nm absorbance/micron determined is -0.015/micron. The 248 nm absorbance/micron determined is 0.07/micron.

EXAMPLE 28 - PMD/TFE A 1 ounce glass Sample vial equipped with rubber septum was flushed with nitrogen, chilled on dry ice, and then 3 ml of perfiuoro(2-methylene-4- methyl-l,3-dioxolane) and 0.5 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃ injected. The vial was set in water ice and allowed to warm slowly to room temperature over the next several hours with magnetic stirring. After -62 hours at room temperature the contents of the vial, a stiff foam, were dried under pump vacuum and then for 17 hours in a 150°C vacuum oven, giving 4.1 g of white fines after crushing with a spatula. DSC, 10°C/min, nitrogen: Tg = 135°C

Inherent Viscosity, hexafluorobenzene, 25°C = 0.155 dL/g

Solution preparation: Rolling 2 g of poly(PMD) with 18 g hexafluorobenezene gave a solution that was filtered through an 0.45 μ glass fiber syringe filter (Whatman, Autovial™). The filtrate was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of PMD were spin coated onto CaF₂ substrates to produce polymer films of 14,818 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for PMD versus wavelength lambda (λ) in units of nanometers is shown in Figure 17 for Sample 33. The 157 nm absorbance/micron determined is 0.603/micron. The 193 nm absorbance/micron determined is -0.0007/micron. The 248 nm absorbance/micron determined is -0.0001/micron.

EXAMPLE 29 - PMD/PDD A 1 ounce glass sample vial equipped with rubber septum was flushed with nitrogen, chilled on dry ice, and then 3 ml of perfluoro(2-methylene-4- methyl-l,3-dioxolane) (PMD), 3 ml of perfluorodimethyldioxole (PDD) and 0.5 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃ injected. The vial was set in water ice and allowed to warm slowly to room temperature over the next several hours with magnetic stirring. After -62 hours at room temperature the contents of the vial, a stiff foam, were dried under pump vacuum and then for 17 hours in a 150°C vacuum oven, giving 8.6 g of white fines after crushing with a spatula. DSC, 10°C/mm, nitrogen: Tg = 147°C Solution preparation: Rolling 2 g of poly(PMD/PDD) with 18 g hexafluorobenzene gave a somewhat gelatinous partial solution that was passed through a ~l/8th inch deep pad of chromatographic silica gel in an 0.45 μ glass fiber syringe filter (Whatman, Autovial™). A portion of this solution was sent for ¹⁹F NMR which found the composition of the dissolved polymer to be 50 mole % PMD and 50 mole % PDD. The remaining filtrate was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of PMD:PDD were spin coated onto CaF₂ substrates to produce polymer films of 12,762 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for PMD:PDD versus wavelength lambda (λ) in units of nanometers is shown in Figure 17 for Sample 34. The 157 nm absorbance/micron determined is 0.404/micron. The 193 nm absorbance/micron determined is 0.006/micron. The 248 nm absorbance/micron determined is -0.002/micron.

EXAMPLE 30 - PMD/VF2 A 75 ml autoclave chilled to <-20°C was loaded with 5 ml of -0.17 M DP in CF₃CFHCFHCF₂CF₃, 15 ml of CF₃CFHCFHCF₂CF₃, and 11.6 ml of perfluoro(2-methylene-4-methyl-l,3-dioxolane) (PMD). The autoclave was pressured with 100 psi of nitrogen and evacuated ten times and then further loaded with 9.6 g of vinylidene fluoride (VF2). Shaking overnight at room temperature gave white solid. Evaporation under nitrogen, 16 hours under pump vacuum, and 32 hours in a 77°C vacuum oven gave 23 g of resin. Calc. (C₅F₈0₂)₁(C₂H₂F₂)₂: 29.05% C 1.08% H Found: 28.71% C 1.38% H

Solution preparation: Rolling 1 g of poly(PMD/TFE) with 19 g of hexafluorobenzene gave partial solution that was passed through a -1/4" deep pad of chromatographic silica in an 0.45 μ glass fiber filter (Whatman, Autovial™). A portion of this solution was sent for ¹⁹F NMR which found the composition of the dissolved polymer to be 54 mole %PMD and 46 mole % vinylidene fluoride. The remaining filtrate was used to spin coat thick films on optical substrate for absorption measurements.

SAMPLE 11 - PPD/CTFE Poly(perfluorodimethyldioxole/chlorotrifluoroethylene), poly(PDD/CTFE), has been reported many times in the literature as for example in US 4,754,009. The poly(PDD/CTFE) used here was prepared by aqueous emulsion polymerization.

Calc. (C₅F₈O₂)₁₀(C₂F₃Cl)₂3 22.52% C 15.93% Cl Found 22.41% C 15.94% Cl

Solution preparation: Rolling 2.5 g of poly(PDD/CTFE) with 30 ml of hexafluorobenzene gave a pale yellow solution that was passed through an 0.45 μ glass fiber filter (Whatman, Autovial™). Because this solution was too viscous to easily spin coat, 11.2 g of this solution was diluted down with another 10.1 of hexafluorobenzene. This diluted solution was used to spin coat thick films on optical substrate for absorption measurements.

Solutions of 10:23 PDD:CTFE were spin coated onto CaF₂ substrates to produce polymer films of 1903 angstroms thickness. VUV absorbance measurements were then used to determine the absorbance per micron.

The absorbance in units of inverse microns for 10:23 PDD:CTFE versus wavelength lambda (λ) in units of nanometers is shown in Figure 18 for Sample 11. The 157 nm absorbance/micron determined is 1.44/micron. The 193 nm absorbance/micron determined is 0.018/micron. The 248 nm absorbance/micron determined is 0.046/micron.

Claims

CLAIMS What is claimed is:

1. A vacuum ultraviolet (VUV) transparent material exhibiting an absorbance/micron (A/micrometer) < 1 at wavelengths from 140-186 nm comprising amorphous vinyl homopolymers of perfiuoro-2,2-dimethyl- 1 ,3- dioxole or CX₂=CY₂, where X is -F or -CF3 and Y is H, or amorphous vinyl copolymers of perfluoro-2,2 -dimethyl- 1,3 -dioxole and CX₂=CY₂.

2. The vacuum ultraviolet (VUV) transparent material of Claim 1, further comprising 0 to 25 mole % of one or more monomers CR^aR^b=CR^cR^d in the case where the CR^aR^b=CR^cR^d enters the homopolymer or copolymer in approximately random fashion, where each of R^a, R^b, and R^c is selected independently from H or F and where R^d is selected from the group consisting of -F, -CF₃, -OR_f where R_f is C_nF_2n+1 with n = 1 to 3, and -OH (when R^c =H).

3. The vacuum ultraviolet (VUV) transparent material of Claim 1 , further comprising 40 to 60 mole % of one or more monomers CR^aR^b=CR^cR^d in the case where the CR^aR^b=CR^cR^d enters the homopolymer or copolymer in approximately alternating fashion, where each of R^a, R^b, and R^c is selected independently from H or F and where R^d is selected from the group consisting of -F, -CF₃, -OR_f where R_f is C_nF_2n+1 with n = 1 to 3, and -OH (when R^c =H).

4. A vacuum ultraviolet (VUV) transparent material exhibiting an absorbance/micron (A/micrometer) < 1 at wavelengths from 140-186 nm comprising amorphous vinyl copolymers of CH₂=CHCF₃ and CF₂=CF₂; CH₂=CFH and CF₂=CFC1; CH₂=CHF and CClH=CF₂.wherein the ratio of monomers ranges from approximately 1:2 to approximately 2:1; perfluoro(2- methylene-4-methyl- 1 ,3-dioxolane) and perfluoro(2,2-dimethyl- 1 ,3-dioxole); perfluoro(2-methylene-4-methyl-l,3-dioxolane) and vinylidene fluoride in any ratio that gives an amorphous composition; and the homopolymer of perfluoro(2- methylene-4-methyl- 1 ,3-dioxolane).

5. The UV transparent material of Claims 1 to 4 wherein the transparent material exhibits A/micrometer < 0.8 at wavelengths from 140 to 186 nm.

6. The UV transparent material of Claim 5 wherein the UV transparent material exhibits A/micrometer < 0.2 at wavelengths from 150 to 160 nm.

7. The UV transparent material of Claim 6 wherein the UV transparent material exhibits A/micrometer < 0.1 at wavelengths from 155 to 159 nm.

8. The UV transparent material of Claim 1 comprising a homopolymer of perfluoro-2,2-dimethyl-l,3 -dioxole.

9. The UV transparent material of Claim 8 wherein the homopolymer of perfluoro-2,2-dimethyl-l,3-dioxole exists as a layer or film greater than 250 nm in thickness.

10. The UV transparent material of Claim 1 comprising a copolymer of perfluoro-2,2-dimethyl-l ,3-dioxole and one or more monomers CX₂=CY₂ wherein X is -F, or -CF₃ and Y is H.

11. The UV transparent material of Claim 10 wherein the copolymer is a copolymer of perfluoro-2,2-dimethy 1-1, 3 -dioxole and vinylidene fluoride.

12. The UV transparent material of claim 2 wherein the polymer is an approximately random copolymer consisting of > 75 mole % perfluoro-2,2- dimethyl-l,3-dioxole and one or more monomers of the structure CR^aR^b=CR^cR^d where each of R^a, R^b, and R^c is selected independently from H or F and where R^d is selected from the group consisting of -F, -CF₃, -ORf where Rf is C_nF_2n+ι with n = 1 to 3, and -OH (when Rc =H).

13. The UV transparent material of Claim 12 wherein the copolymer is

> 75 mole % of perfluoro-2,2-dimethyl- 1,3 -dioxole and < 25 mole % tetrafluoroethylene.

14. The UV transparent material of Claim 3 wherein the polymer is an approximately alternating copolymer having 40-60 mole % CX₂=CY₂ where X is -F or -CF₃, and Y is H, and having 60 to 40 mole % monomers of the structure CR^aR^b=CR^cR^d where each of R^a, R^b, and R^c is selected independently from H or F and where R^d is selected from the group consisting of -F, -CF₃, -ORf where Rf is C_nF_2n+1 with n = 1 to 3, and -OH (when R^c =H).

15. The UV transparent material of Claim 14 wherein the copolymer is a hexafluoroisobutylene/trifluoroethylene copolymer wherein the ratio of the monomers is from approximately 60:40 to approximately 40:60.

16. The UV transparent material of Claim 14 wherein the copolymer is a hexafluoroisobutylene/vinyl fluoride copolymer wherein the ratio of monomers is from approximately 60:40 to approximately 40:60.

17. The UV transparent material of Claim 2 wherein the copolymer is an approximately random copolymer having > 75 mole % CX₂=CY₂ where X is -F or -CF₃, and Y is H, and having <25 mole % monomers of the structure CR^aR^b=CR^cR^d where each of R^a, R^b, and R^c is selected independently from H or F and where R^d is selected from the group consisting of -F, -CF₃, -ORf where Rf is C_nF_2n+1 with n = 1 to 3, and -OH (when R^c =H).

18. The UV transparent material of Claim 17 where the copolymer is >75 mole % vinylidene fluoride and <25 mole % hexafluoropropylene.

" 19. The UV transparent material of Claim 14 wherein the copolymer is 60-40 mole % vinyl fluoride and 40-60 mole % chlorotrifluroethylene.

20. Pellicles comprising the UV transparent material of Claims 1 to 4.

21. Anti-reflective coatings comprising the UV transparent material of Claims 1 to 4.

22. Optically clear glues comprising the UV transparent material of Claims 1 to 4.

23. Light guides comprising the UV transparent material of Claims 1 to 4.

24. Resists comprising the UV transparent material of Claims 1 to 4.

25. The resists of Claim 24 further comprising solubility responsive monomers.

26. Transmissive optical elements comprising the UV transparent material of Claims 1 to 4.

27. A copolymer composition comprising poly(hexafluoroiso- butylene:trifluoroethylene) with 40-60 mole % hexafluoroisobutylene and 60-40 mole % trifluoroethylene.

28. An amorphous copolymer composition comprising poly(hexafluoroisobutylene:vinyl fluoride) with 40-60 mole % hexafluoroisobutylene and 60-40 mole % vinyl fluoride.

29. A copolymer composition of Claim 27 or Claim 28 having A/micron

< 1 at 140-186 nm.

30. A copolymer composition of Claim 29 having A/micron < 0.8 at 140-186 nm.

31. A copolymer composition of Claim 30 having A/micron < 0.2 at 150-186 nm.

32. A copolymer composition of Claim 31 having A/micron < 0.1 at 155-186 nm.

33. A process for the preparation of films of PDD having thickness greater than 250 nm comprising the steps of: a) placing monomer, optionally diluted with solvent and/or initiator, in the location where the film is desired, and b) initiating poymerization by physical and/or chemical means.

34. The use of the polymer of Claim 18 as an adhesive.

35. The use of the polymer of claim 18 as an adhesive in attaching a pellicle to a photomask.

36. The use of the polymer of Claim 18 as an adhesive in attaching a pellicle to a frame.